Myths of Evolution (4): The Selfish Gene

In 1976, Richard Dawkins published his book The Selfish Gene, which built upon George C. Williams ground-breaking idea that “adaptation” was too vague a term to build a scientific theory upon, and that evolution would be better understood as selection among genes or individuals. Dawkins helped develop this concept into what is now called the gene-centric view. This perspective (or model) provides a valuable view of the history of life and one cannot truly claim to have grasped modern theories of natural selection without some appreciation for how things seem when viewed from the perspective of genes.

However, as useful as this model can be it is rife with misconceptions. The mythologies attached to “the selfish gene” can be broken into two distinct camps: firstly, erroneous iconography created by the wording of this phrase when the individual does not understand the gene-centric viewpoint Dawkins was espousing, and secondly, an overly reductionist dogma among some scientists who mistake an explanatory principle for a fundamental law.

The problem with the idea of a “selfish gene” in popular culture is that what 'selfish' means to most people is utterly different to what 'selfish' means in the gene-centric view. This criticism is meticulously developed by the philosopher Elliott Sober in his essay What is Evolutionary Altruism? which demonstrates that the way “selfish” and “altruistic” are deployed by evolutionary biologists is radically different from the way these terms are used colloquially. He uses the example of giving someone a piano: this can be altruistic in the vernacular sense (depending upon the motive behind the gift), but in the sense used in evolutionary theory, the piano may distract you from having babies , thus reducing your “evolutionary fitness”.

Arguably more pernicious is the dogmatic belief in an ideology derived from the gene-centric view. Stephen Jay Gould and Richard Dawkins fought bitter public battles over their differing interpretations of evolutionary mythology, and Gould was especially scathing about Dawkins' obsession with genes, claiming that the gene-centric view was “a confusion of bookkeeping with causality”, providing an overly reductionist perspective which he characterised as “Darwinian fundamentalism”. Gould's criticisms in this regard arguably overstepped the mark, probably as a result of the anger and hostility engendered by the heated conflict between the two highly regarded scientists.

The problems with predicating the gene-centric view as a fundamental principle are multifarious, but a brief but key complaint is that genes do not generate behaviour in and of themselves, and much of Dawkins and others reasoning in this regard depends upon this connection. Genes are DNA code for proteins, and although these proteins are used to construct behavioural systems, such as elements of the brain, neurotransmitters and hormones, the gene is just a component of the template for the organism's biology. Just as a brick is used to construct a building but tells you little about what people do in buildings, a gene helps build a body but by itself tells you little about what that body does. Behaviour – even among simple animals – depends as much on environment and culture (or ecology for less complex lifeforms) as it does the biological capacities inherited via genetic transmission.

Rather than “the selfish gene” – which risks misrepresenting the natural history of life as driven solely by miserly competition (a myth exposed last week) – the gene-centric view can perhaps be better grasped in common parlance by the idea that advantages persist. This is indeed the essence of the gene-centric view: a gene that leads to advantages for an individual (and by extension, a species) is vastly more likely to persist, and this persistence of advantage is the ratcheting mechanism that drives, in another of Dawkins' metaphors, the ascent up the slopes of “mount improbable” - the landscape of all possible life.

Alternative myth: Advantages Persist

Next Week: Myth #4: Kin Selection

Comments

[Peter Crowther]

I love the opening image - superb!

I disagree strongly that an acceptable alternative for the gene-centric view is that "advantages persist"; that broadens the statement to include many non-gene-centric views, clouding the issue and reducing the testability of any of the various views. An "advantage" is unlikely to be the result of a single gene. The organism typically needs combinations of (possibly non-adjacent) genes to be transferred together to its offspring to preserve the advantage - one of the key problems with the gene-centric view in general, and one of its key tests.

All of my response above assumes that "gene" is approximately a shorthand for "adjacent series of base pairs in an organism's DNA". If we are playing different language games, this statement may help to clarify that.

[William Monroe]

I consider myself spectacularly unqualified to hold much of an opinion on Evolutionary Biology, but I found your reference to the heated debate interesting.

Funny how our accomplishments and identities can become conflated, and an attack on one can seem like an attack on the other.

[Chris]

Peter: well remember I'm not suggesting the gene-centric view is best encapsulated in the phrase "advantages persist", I'm suggesting that as a mythology to give an everyday interpretation of the consequences of the gene-centric view "advantages persist" is as valid (or more valid) than "the selfish gene". A DNA coding isn't selfish in any meaningful way, and the consequences of gene-centrism is not selfish behaviour - I consider this spin to be quite misleading.

I also think that "advantages persist" works as a spin on the teleological stories that people like to tell using the gene-centric view, which almost never pertain to individual genes. In fact, it's a struggle to find instances of individual genes yielding selective advantage - which is fundamental to how many Neo-Darwinists field this model. The glycoprotein in antarctic cod springs to mind, or camouflage colours in insects perhaps. Most of the really big advantages require more than one gene.

Plus, a key point of my perspective here is that the genes by themselves aren't enough - it's also about the environmental circumstances, which includes the interactions with other species. In the case of camouflage colours in insects, for instance, it's not the gene that renders the advantage, it's the outcome of the gene in concert with the specific predatory relationships in the local ecology.

The central idea of the gene-centric view is that those alleles that have phenotypic effects that are beneficial to the organism have a selective advantage over those alleles that do not, and are thus increase in frequency. But those phenotypic effects - as you say - are not the result of single genes. So it's not clear to me that the gene-centric view doesn't already need to be considered in a context wider than the individual genes.

I think "advantages persist" is a sound myth to tell about evolution, one which the gene-centric view contributes towards. But it is not intended as a one-to-one correlation. It's also the essence of Dawkins central metaphor in "Climbing Mount Improbable" (as I allude to here) which I still feel is one of the best expressions of why evolution can create such amazing complexity and diversity of life.

William: I found the conflict between Gould and Dawkins to be fascinating precisely for the reasons you allude to here - it went beyond the scientific disputes and into personal attack. No matter how intellectual we may be, our emotional responses are can still be quite raw and often volcanic! :)

[Peter Crowther]

A little rushed for time, so I'll pick a few phrases and comment. I hope these clarify my own position and the lanuage isn't inflammatory!

A DNA coding isn't selfish in any meaningful way

... what function are you using to assign meaning?

the consequences of gene-centrism is not selfish behaviour

I think that depends on your point of view, both linguistically and philosophically :-). I also suspect we're into metaphysics here, particularly with modifications to the theory that allow for decreasing the chance of survival to reproduction of one piece of gunk that happens to carry a gene in exchange for increasing the chance(s) of survival to reproduction of other gunk that may carry the same gene. For example, it's unlikely that I will reproduce. It's likely that I will (marginally) increase the chances of survival to reproduction of some other humans, who share my DNA to a greater or lesser extent. I'd love to see the analyses of behaviour towards others based on the expected commonality in DNA - both in environments that were resource-constrained and those that were not.

Most of the really big advantages require more than one gene.

Yes, and it's fascinating (as part of my work) to see how many of the small ones do!

This requires an explanation of why bundles of functionality that's coded by non-adjacent pieces of genetic material appear to persist in the population. My own explanation is around the laws of large numbers and long time periods. The evolutionary playing field is tilted at a very shallow angle, and advantages are generally very minor. At this point, you can have a remarkably inefficient system for transferring advantages around the population and it still works. Humans are wired to understand small numbers and human time periods, and I think many people don't "get" evolutionary theory because they have no feel for the sheer numbers of individuals and generations involved.

a key point of my perspective here is that the genes by themselves aren't enough - it's also about the environmental circumstances

Yes. A "gene for" anything may be beneficial, neutral or harmful to any individual in the environment in which that individual finds themselves. Note the scare quotes to acknowledge that one set of adjacent base pairs is rarely sufficient.

But... I neither understand why this is news, nor why it invalidates the gene-centric or selfish-gene viewpoints. It seems entirely compatible. Could you point me to a source that describes the incompatibility in more detail, as I confess I can't see it.

I think "advantages persist" is a sound myth to tell about evolution

I agree. But I don't think it is an alternative to the selfish gene.

[Isegoria]

"Advantages persist" conveys the notion of evolution, but it misses the key insight of "the selfish gene": that selection operates not on the individual, nor on the species, but on the genetic code.

It's painfully literal to assert that DNA "isn't selfish in any meaningful way," when it obviously isn't consciously selfish, but it does propagate if it leads organisms to do what's in its "interest" (the gene's, not the organism's).

It's also painfully reductionist to claim that genes don't matter, because they don't do anything "by themselves" — as if anything does. Of course a gene operates within a context. The "fittest" in "survival of the fittest" implies a context to fit into.

[Chris]

Peter: I enjoyed your comment here, but I don't have much to respond to. I would check the Elliot Sober essay if you haven't already, for discussion of the meaning of "selfish".

Incidentally, are you aware that recent research has shown that genes for various things previously assumed to have independently evolved are actually shared by almost everything that have that feature? For instance, the PAX6 gene in eyes. Now it is a matter of debate, but if we all share the PAX6 gene - from fruit flies to mammals - how an earth does the "selfish gene" deal with this datum?

The idea that one gene can have a "monopoly" across all species suggests that conference of advantage is more important than competition between distinct alleles, at least in my interpretation.

But this is on the fringes of research, and the room for debate is huge!

"I think many people don't 'get' evolutionary theory because they have no feel for the sheer numbers of individuals and generations involved."

I agree! But I also think that those of us who are not bamboozled by "the dark abyss of time" sometimes feel we understand everything about evolutionary theory simply because we have a handle on something that most people do not.

Isegoria: "...but it misses the key insight of 'the selfish gene': that selection operates not on the individual, nor on the species, but on the genetic code."

This is a teleological account you (and others) are applying at a metaphysical level. Selection operates on the individual AND the species AND the genes. This is precisely Gould's critique of Dawkins - that he brought back the teleology that Darwin was supposed to have expunged - and in this regard Gould was completely accurate.

And while we're at it, selection operates at the level of the ecology as well. Professor Armand Marie Leroi suggests we might want to start thinking of the evolution of *systems* (rather than organisms/species) and I support this interpretation.

And I'm not claiming genes don't matter. Why on earth would you think that? I'm just saying you can't reduce the evolutionary process solely to genes - except by introducing teleology at a metaphysical level. The meaning of the genes changes entirely according to environment and ecology. The same genes in a plant produce entirely different plants under different conditions, as R.C. Lewontin demonstrated.

Thanks for your comment!

[Isegoria]

This is a teleological account you (and others) are applying at a metaphysical level.

"You keep using that word. I don't think it means what you think it means." — Inigo Montoya

I find your use of teleological puzzling, because the key insight of evolution is the antithesis of teleology: we can get "progress" that appears to be goal-directed despite a lack of any such goals.

In your previous piece on telelogical games, you raised a legitimate point that many evolutionary "just so" stories aren't particularly scientific — after the fact, we simply say, "this changed, so it must have made the organism more fit."

But evolutionary science can and does make predictions that come true — and flawed facets of older versions of evolutionary science have been discarded, because they did not fit the facts.

A naive organism-centric view of evolution — "survival of the fittest" — explains many, many things, but it doesn't predict altruism towards kin the way a gene-centric view does, for instance — which the gene-centric view predicts quantitatively.

And while we're at it, selection operates at the level of the ecology as well.

Ecologies can change and evolve without selection working on the level of ecologies.

The key insight of the gene-centric view is that it's the genes that replicate or fail to replicate, so selection can only work on the level of the genes — but these genetic changes show up in individuals, which make up species, which are part of larger ecosystems, etc.

[Chris]

Isegoria: You're right of course that the whole point of the explanatory success of evolutionary theories (there isn't just one) was to remove the teleological element from the equation. But that's precisely why you have to be wary about accidentally popping it back in. :)

The best thing I can recommend is to read Gould's criticisms of Dawkins' position. I can't find the one in which he levels the accusation that "the selfish gene" risks reintroducing teleology, but you can get an aspect of this from his New York Review of books article from the height of Gould's public battle with Dawkins et al. Gould is consequently rather miffed - and understandably so if you read what his opponents have levelled against him at this point! I think this piece goes too far, but its central points are still relevant.

You clearly believe that what is valuable about the gene-centric view is that selection can only work on the level of the genes. This is your interpretation, which you are entitled to, and you share it in common with Dawkins, Dennett etc. But it's a myth of evolution - a story spun from the scientific models. Note that "myth" doesn't mean false - it just means it's outside of the strictly testable world of science.

Without a strict account for behaviour in terms of genetics (which has not been found, and now looks doubtful) you cannot account for everything in biology solely in genetics. The prevailing view of the evolutionary scientists outside of the strict adaptionist camp is towards more multi-faceted approaches (what Gould calls "pluralist" approaches).

If you are not familiar with Kimura's neutral theory, which Gould mentions in the piece I linked to above, I heartily recommend looking into this as this is a vital piece in the evolutionary puzzle. Alas, I don't have time to dig into it here, but I recommend this as a point of enquiry. If you like, I can put up a piece about this at some point as it could make for some interesting reading if I can find a not-too-technical way to present it. (This has been what has stopped me thus far).

You can't build ecologies solely from genes - the genes don't explain which species are in any given place at any given time, any more than they completely explain behaviour, and it is the combinations of species which more than anything dictate the prevailing conditions in any biome. If you are not happy using 'selection' for anything other than 'gene selection', fine, but this choice doesn't mean you can eliminate the other factors in the evolutionary story without falling into a teleological trap of some kind.

We are a long way still from a unified account of evolution - Gould's untimely death has been a setback in this regard, as the voices in the camp opposed to strict adaptionism (being less prominent/famous now than Dawkins) have not been in a position to renew the debate publicly. But this absence of public discussion does not equate to an absence of dispute - I assure you, if you dig into it, you will find considerable philosophical conflicts throughout every field that neighbours on evolutionary research.

Hope this addresses your concerns, and if not I'm happy to discuss this further!

[zenBen]

"the combinations of species which more than anything dictate the prevailing conditions in any biome"

Funny, I thought it was the prevailing conditions in the biome which dictated the combinations of species!

http://www.world-science.net/othernews/090204_titanoboa

Or, maybe its just a complex chaotic mix of the two.

[Chris]

zenBen: you're right, of course - it's a chaotic blend of both factors.

Climate and soil conditions determine which plants can grow (and the evolutionary space plants can diversify within), and the rest of the species form a highly involuted and interconnected web on top of this.

But you certainly can't look at the inorganic conditions and deduce the combination of species in the biome (although you can, as in your link, observe features of species in the biome and explain aspects in terms of inorganic factors).

And again, it becomes quite circular, since the species are also influencing the inorganic conditions: lichens break down rock into soil, the albedo of plants changes the reflectivity of an area affecting the temperature, the ratio of gas exchange across all species affects climate and weather...

The idea that environment and organism are easily separated can be quite misleading!

[Isegoria]

The best thing I can recommend is to read Gould's criticisms of Dawkins' position. I can't find the one in which he levels the accusation that "the selfish gene" risks reintroducing teleology, but you can get an aspect of this from his New York Review of books article from the height of Gould's public battle with Dawkins et al. Gould is consequently rather miffed - and understandably so if you read what his opponents have levelled against him at this point! I think this piece goes too far, but its central points are still relevant.

Frankly, I found that piece embarrassing to read, with its empty ad hominem attacks — ultras, fundamentalists — and straw men — like this gem:

Natural selection does not explain why a meteor crashed into the earth 65 million years ago, setting in motion the extinction of half the world's species.

I almost choked when he said:

A false fact can be refuted, a false argument exposed; but how can one respond to a purely ad hominem attack?

But I haven't read the attacks on Gould; perhaps they were as empty.

You clearly believe that what is valuable about the gene-centric view is that selection can only work on the level of the genes. [..] But it's a myth of evolution - a story spun from the scientific models. Note that "myth" doesn't mean false - it just means it's outside of the strictly testable world of science.

You said this before, and you seem to be ignoring my counter-point that the selfish gene is not outside the testable world of science. Scientists have made predictions based on the gene-centric model that came true when tested. For instance, the level of sacrifice an organism is willing to make for kin should depend on the fraction of genes shared, and it does indeed seem to.

Without a strict account for behaviour in terms of genetics (which has not been found, and now looks doubtful) you cannot account for everything in biology solely in genetics.

You're propping up a straw man. No one claims that you can account for everything in biology solely via genetics. The claim is that genes are what is inherited; thus selection operates on genes.

And, really, anything that gets transmitted like genes can undergo natural selection. That's the whole point of Dawkins' memes.

Lastly, you definitely do not need a strict account for behavior in terms of genetics. All you need to know is that genes can account for behavior at all. Then you know that something that can be passed along and selected for can shape behavior.

I think it's a mistake to attack the selfish gene for failing to explain all of nature, when it succeeds at explaining so many things that couldn't be explained before.

[Chris]

Hi Isegoria, there seems to be some confusion here as to how I have used "selfish gene" versus "gene-centric view". I don't consider these terms to be interchangeable - "selfish gene" is a specific spin on the gene-centric view in my use of the term. When I mean gene-centric view, I write this. When I write "selfish gene" I mean the spin, not the theory it is supposed to refer to, that's why I call it untestable.

I attack "the selfish gene" as a misleading spin on the science. The gene-centric view, I have no issue with, and agree with you that it explains a lot of issues that previously lacked an explanation.

"No one claims that you can account for everything in biology solely via genetics."

Are you sure?

"The claim is that genes are what is inherited; thus selection operates on genes."

But genes are not the only things that are inherited, there is also epigenetic programming and non-genetic information such as culture and its non-human equivalents. Not to mention that the composition of species in a particular biome is not a genetic function, but something entirely different.

"And, really, anything that gets transmitted like genes can undergo natural selection. That's the whole point of Dawkins' memes."

Yes, but cultural information *isn't* transmitted like genes. The whole reason that genetics explained a lot of the problem areas in evolution at the start of the twentieth century was that genes were not blended (as previous inheritance mechanisms had presupposed) but discretely inherited - but cultural information *is* blended, and is not linked to single underlying inheritable elements. It is not at all like genetics. This is why the meme proposition dries up quite rapidly, as most people in and around this field now recognise.

As for the Gould piece, you should really read all the exchanges to get the full picture here... everyone was clearly pretty steamed at this point, and it shows.

[Isegoria]

When I write "selfish gene" I mean the spin, not the theory it is supposed to refer to, that's why I call it untestable.

I still don't see the "spin" here. That's why I'm confused. How does the selfish gene distort the gene-centric view of evolution?

Are you sure?

Yes, I am sure that no one (sane) claims that you can account for everything in biology solely via genetics.

If you put an animal in a cage and starve it, or if you shoot it through the heart, no one is going to claim that the animal died solely or even primarily because of its genetics — unless there's a good case that you picked that animal for genetically influenced reasons, I suppose. Humans have, apparently, shaped prey species — non-domesticated ones — by killing the largest males for trophies. But that's quite peripheral to the larger point that we all know genetics and environment interact. The key is that genes are heritable and reproducible.

But genes are not the only things that are inherited, there is also epigenetic programming and non-genetic information such as culture and its non-human equivalents.

To the extent that those things are heritable, they can presumably evolve. Culture obviously evolves, if not by the same mechanisms as genes encoded in DNA, and culture and genes can shape one another, in a species like Homo sapiens. I don't dispute that, and I doubt Dawkins et al. would either. (In fact, I think it's Gould, author of The Mismeasure of Man who would likely have objections, because of the implications.)

Yes, but cultural information isn't transmitted like genes.

I suppose that depends on our definition of like. It is transmitted, it does mutate along the way, and it does face various selection pressures, but the analogy isn't perfect, we can agree. (I don't think I'd go so far as to claim that it's blended though.)

As for the Gould piece, you should really read all the exchanges to get the full picture here... everyone was clearly pretty steamed at this point, and it shows.

That's what I had to assume.

[Chris]

Isegoria: Ah, you want to restrict your scope to *sane* people. Well, that radically curtails the scope of the discussion. Let me know when you have found the sane person. >:)

It seems like we have exhausted our major points of difference. You still don't see why I feel that presenting the gene-centric view as "the selfish gene" is a distortion, but you recognise that a gene cannot be selfish, and that the gene-centric view in no way means the guarantee of selfish behaviour, so we're in agreement on the fundamentals, I think.

So I suspect the root of the problem here is that because you are an intelligent, learned individual, you can't believe that someone could misunderstand what "the selfish gene" is supposed to entail. Yet people do, all too frequently, misunderstand this and - worse - fall into the teleological trap of making statements like "my evolutionary purpose is to pass on my genes".

But clearly "evolutionary purpose" is some kind of teleological claim. You pass on your genes if you breed, but it is not your *purpose* (evolutionary or otherwise) to do so, at least, not without asserting teleology a priori to the situation. The most we can say is that our biology strongly encourages us to have sex (but, following Bagemihl, not even necessarily to breed!), but that falls radically short of a claim of purpose. Science in my view shouldn't be making this kind of metaphysical claim, nor encouraging them, and that's the root of my objection here.

Enjoyed chewing this over with you - your objections helped to clarify my arguments considerably, and that's what a good debate should do in my opinion.

Thanks again!

[Peter Crowther]

Science in my view shouldn't be making this kind of metaphysical claim, nor encouraging them, and that's the root of my objection here.

Chris, in previous pieces you have carefully distinguished science from scientists. In this comment, do you mean Science, or do you mean those fallible, all-too-human scientists? If the latter, do you mean all, most, some or a few?

[Chris]

Peter: you are ever-ready to catch these careless mistakes in my prose. You are right, I should probably have said "scientists" not "science" (although by "science" I mean "the actions and models of scientists", so...) But once this is identified, the whole sentence needs re-writing:

"Scientists are free to make metaphysical claims, as indeed are we all, but in their role as advocates of the scientific endeavour I believe scientists should avoid encouraging one metaphysical interpretation over another."

Does that resolve your objection?

[Peter Crowther]

Yes, that resolves my objection - although it does put some very strict constraints on the more prominent advocates of the scientific endeavour, as it is very hard to prevent reporting of a scientist's metaphysical views in the form "Dawkins, the noted scientist, believes...".

OK, next question - I'm not as widely read on the subject as you are. Which scientists, in their role as advocates of scientific endeavour, 'fall into the teleological trap of making statements like "my evolutionary purpose is to pass on my genes"'?

[Chris]

Peter: if scientists were better trained in philosophy of science they would be able to distinguish where their personal metaphysics were involved in the presentation of theoretical or hypothetical materials. ;)

But re: "my evolutionary purpose...", it's not the scientists who tend to make this mistake - it's the general public, having misunderstood what the scientists are advancing. I don't think the scientific community necessarily misunderstands "the selfish gene", but the general public (from my experiences talking to people about their understanding of evolution) frequently does.

In the case of Dawkins, whose very role in the end was supposed to be a professor for the public understanding of science, these kinds of widespread misunderstandings represent some kind of failure - which is not in any way to denigrate what Dawkins was able to contribute to public understanding. (I myself learned a lot from him). It is just an admission that we must be careful when we present theories to the public using just a snappy phrase to bridge the gap of comprehension.

Best wishes!

[Peter Crowther]

if scientists were better trained in philosophy of science they would be able to distinguish where their personal metaphysics were involved in the presentation of theoretical or hypothetical materials.

And the "popular" press would still deliberately distort the carefully presented words for the express purpose of selling more newspapers / getting more eyeballs. Dissemination of science via any means (such as arts graduates without science O-levels becoming primary school teachers) is a horribly noisy channel, and I no longer believe that we are capable of doing it effectively as a society.

More on the substance of the above when I don't have to run to work!

[Jerzy Wieczorek]

Hi! I'm trying to understand what it is that the Dawkins and Gould camps are arguing about, and googling led me here -- very interesting post and comment thread! Do you mind if I ask for clarification?

It sounds like Dawkins thinks a gene-centric view of evolution (i.e. it's the genes that are "selfish," not individual organisms or species) is better at explaining many things that are problems for the organism- or species-centric view.

As far as I can tell, neither you nor Gould disagrees with that part. Gould just gets angry when people oversimplify and say gene-centrism is ALL you need to explain the diversity of life, since...
(1) ...that implies that it's okay to look at only genes while ignoring the context of the environment. (But of course Dawkins knows a gene's success depends on the context of the other genes in the same body as well as the environment and ecology in which that body happens to live.)
(2) ...some features of life aren't adaptations honed by natural selection, but perhaps arbitrary biological features that don't provide a significant cost OR benefit to their bearers and thus don't fall into the category of Things That Gene-Centric Natural Selection Can Sensibly Be Used To Explain. (That's a reasonable criticism -- evolutionary psychology in particular seems to get carried away with this.)
(3) ...genes don't directly cause behavior, they only code for proteins that start a chain of biochemical reactions that blah blah etc. (That argument seems hollow to me... For example, sure, there's no "gay gene", but presumably the general behavior of mating has some genetic basis!)
(4) ...selection CAN happen on things other than genes, for example on species and even on ecologies.

Arguments (1-3) seem to be Gould's frustration with people who oversimplify or overuse gene-centrism. Right on. But (4) seems directed at Dawkins etc directly, and it's the argument I don't understand at all. Could you explain it again without the word "teleology"? I thought it refers to intentional, purposeful design. I utterly fail to see why you think gene-centrism brings intention and purpose back into evolutionary theory.
Gene-centrism seems to be presented as a "better" view of evolution NOT because the genes have some "intent" of their own, but in the sense that genes (or gene complexes) are precisely the thing that gets replicated. You just plain can't replicate an organism or an ecology, only their genes. So there is a way in which it just plain doesn't make sense to talk about evolution acting on individuals or species or whatever. Yes, ecologies change over time, but in a qualitatively different way than genes and gene complexes do. What does that have to do with teleology?

Thanks and best wishes!

[Jerzy Wieczorek]

After another reading it maybe makes a bit more sense... You're not saying the "gene-centric view" is teleological if correctly understood; you're saying that there is a particular misinterpretation of it (which you call "the selfish gene") that is teleological, and that misinterpretation includes "My evolutionary purpose is to pass on my genes."
Whereas a good Dawkins-side scientist should only say "We cannot talk of purpose; and sadly it's impossible to explain many evolutionary developments in hindsight because we have a limited fossil record and no way to rerun history experimentally; BUT in situations where testable hypotheses are possible, the gene-centric view seems to be a more sensible way of proposing hypotheses than organism-centric, species-centric, or ecology-centric views would be."

Okay. But it's really confusing that you use "selfish gene" as your term for this misunderstanding of Dawkins' idea, since that's the name Dawkins himself gave it! Now that I understand, I can see it... but it took me several readings to realize that that was your point, and I don't think Isegoria got it at first either. This article would have been a lot clearer if you referred to the misinterpretation as "ConfusedSelfishGene" or something like that.

So would you say Dawkins' original presentation falls under the misinterpretation category, or is it just that Gould misread it that way? Or did Gould actually have problems with the proper gene-centric view itself too?

Thank you!

[Chris]

Jerzy: Great comments! Thanks for wading in on this. A few answers to your questions...

"It sounds like Dawkins thinks a gene-centric view of evolution (i.e. it's the genes that are "selfish," not individual organisms or species) is better at explaining many things that are problems for the organism- or species-centric view."

Well Dawkins takes a very hard line here and often seems to be saying that the gene-centric view is all that is required for explanation of evolutionary questions of life. This view is, as far as I can ascertain, fundamentally incorrect. But that doesn't invalidate the gene-centric view. It is only the elevation of that view to the status of exclusive axiom which oversteps the mark.

Your breakdown of Gould's position seems accurate; I think the fight between Dawkins and Gould was less about research and more about dogma and personality. Of your list, (2) was Gould's biggest sticking point with Dawkin's view, and (1) a secondary concern. (3) and (4) are more my concerns than Gould's, I think. My apologies for jamming these all together in one piece! :)

"You're not saying the 'gene-centric view' is teleological if correctly understood; you're saying that there is a particular misinterpretation of it (which you call "the selfish gene") that is teleological, and that misinterpretation includes 'My evolutionary purpose is to pass on my genes.'"

Yes, this is precisely the point! My apologies if this didn't come out clearly in the original piece. I appreciate it's confusing that I use "selfish gene" and not some other distinct term. This is a fair criticism of this piece.

"So would you say Dawkins' original presentation falls under the misinterpretation category, or is it just that Gould misread it that way? Or did Gould actually have problems with the proper gene-centric view itself too?"

Gould (and I) did (and do) have problems with the gene-centric view when it is taken as exclusively axiomatic i.e. that the whole of evolutionary history can be understood *solely* as selection of genes. This isn't enough; one must also allow for the effects of chance that are not the result of directed gene selection (Gould's point) and for persistent elements that affect gene selection e.g. specifics of environment and ecology (my point).

Hope this clarifies!

[Jeena Paradies]

What do you mean by "'directed' gene selection"? Isn't the whole point that there is no one who directs the natural selection (as opposed to artificial selection)?

And then in fact Dawkins mentions "chance" (or randomness) explicitly as one of the fundamental units of natural selection:

"The fundamental units of natural selection, the basic things that survive or fail to survive, that form lineages of identical copies with occasional random mutations, are called replicators."

And if you search the book for the term "environment" you will find tens of examples where he states that it is in fact the environment which affects gene selection via pressure (weather, predators and prey, supporting vegetation and soil bacteria, etc.).

[Chris]

Jeena:

"What do you mean by 'directed gene selection'? Isn't the whole point that there is no one who directs the natural selection (as opposed to artificial selection)?"

No-one, but not no-thing. If there was nothing affecting gene selection we could scarcely call it selection! The environment, the nature of the species within that environment, and the confluence of interactions between that species and other species - collectively, the circumstances of life attending to any organism - provide direction to gene selection, or to put it another way, furnish probability gradients for the persistence of genes.

Gould's point is that this isn't the whole story - the genetic circumstances for any organism depend upon more than just selective pressures, and we cannot assume that any given gene is present solely for selective reasons. There are also random events with significant influence (such as mass extinctions) and also legacy features ('spandrels') which cannot be deduced as a product of selection circumstances. Therefore attempting to explain all features of an organism in this way risks error.

Gould's point concerning chance is different from Dawkins'; Dawkins is talking about chance as per the essential source of genetic novelty - a view argued as incomplete by Margulis and others. This issue of random mutation lies outside the scope of what I talk about here, but it's an interesting one.

Personally, I have advanced a view I call the cut and paste hypothesis whereby these "random" mutations are in fact virus-induced insertions of introns DNA as exons. I suspect geneticists must have advanced the same view, but since genetics is not my field I don't stay on top of the literature to know whom I should cite in this regard. :)

The point of this post (and its attached commentary) isn't that Dawkins' perspective on evolution is fundamentally in error, but just that the characterisation of the history of genetic material as "selfish" in any sense misuses this term in a quite outstandingly bizarre fashion. He is much closer to fair metaphors in his later works than he was with "The Selfish Gene" ("Climbing Mount Improbable" is my personal favourite).

Ironically, this connects with the current Pentenary serial since today's post is entitled The Gene Confusion, although the most pertinent part of the argument in respect of what is going on here won't appear until September 9th in a piece entitled "Myths of Selfishness". If this topic interests you, I encourage you to stick around and get involved in discussions.

Thanks for commenting!

[Jeena]

Thank you for the clarification, but I still think that everything you list can be reduced to pressure, release of pressure within the environment.

For example a random event such as mass extinctions of the dinosaurs releases the (environmental) pressure on mammals and therefore natural selection now can favore bigger animals.

I understand that there are basically three types of mutations: harmful, beneficial, neutral/silent. Environmental pressure affects the individuals who have harmfull or beneficial mutations but not the ones who have neutral mutations.

While researching for the next paragraph where I wanted to show some numbers which would show that neutral mutations are quite uncommon I realized that this can not the case. Then I found out about the "Neutral theory of molecular evolution" and the "neutralist–selectionist" debate. if I understand it right, this post (and its attached commentary) is one of the "neutralist–selectionist" debates.

The "Neutral theory" is a whole new chapter which I do not know anything about but will of course try to read about and understand. Anyway, thanks for opening that door for me and I will definitely stick around and get involved in discussions.

[Chris]

Jeena: Your paradigm in terms of pressure probably holds up, but it is not the only point of view of course. It's worth being aware that any given model only gives you one of many perspectives. There is not necessarily a model with a definitively privileged status in any domain of science.

I recommend you do dig into the neutral theory a little - it's very critical to understanding what's going on in evolutionary studies, and you don't need to dig deep to get at the key point.

The vast majority of mutations of genes result in death for the animal in question, because small changes to a protein configuration destroy its ability to work in its functional role. However, mutations in introns (non-gene) DNA don't have this problem, and Kimura demonstrates that introns DNA shows greater rate of mutation as a result.

My contribution (which I'm sure actual geneticists have also mooted) is that new genes probably result when introns DNA is spliced in as a gene by a virus - this seems to have happened in the case of the glycoprotein in Atlantic cod (I think that's the right fish, anyway). If so, then even random mutation comes about as a result of interactions between organisms (or quasi-organisms, since viruses are on the borderline).

Best wishes!

[Bhengeveld]

Hello Chris, always interesting to read critiques like these. Let me get this straight (I'm no native English speaker and nu scientist): your main point is that the term 'selfish' is misleading because of the way it influences how people think of the gene? (i.e. loading selfish with the moral meaning we usually give to the word, when evolution itself is devoid of it)

Then, you state (with Gould):

"the genetic circumstances for any organism depend upon more than just selective pressures, and we cannot assume that any given gene is present solely for selective reasons"

The way I read Dawkins I think of a gene as a unit that just 'is' and had the opportunity to replicate. Having had this opportunity is in essence 'selection'. (just as 'selfish', 'selection' is perhaps laden with teleological implication that are ill-suited in this context?) Selection doesn't have to happen because it gives the organism a distinct advantage (the organism is there for the gene, not the other way around), so 'pressure' and 'selection' are words that have (very) distinct meaning.

[Bhengeveld]

Sorry, pressed 'Post' instead of 'Preview'. I have to add that if my explanation holds up, we don't have to assume (at all) that "any given gene is present solely for selective reasons".

That would be another wrong interpretation of what a selfish gene is.

[Chris]

Bhengeveld: thanks for your thoughtful comment, and despite your objections your English is very good!

Yes, the problem with "selfish" is that this has a common meaning that implies *motive*. Mary Midgley mounts an excellent objection to this kind of talk for this very reason. It's misleading to use words that imply motive to entities that have nothing of the kind, and we'd be wise to avoid doing so. The people who do so, including Dawkins, may be tacitly touting a perspective on human motivation (e.g. Hobbes) that the theory under discussion has nothing to say about... Dawkins does much better in Climbing Mount Improbable, which in my view is a superior book on this subject.

Your point re: "selection" being potentially misleading is excellent - it is indeed the case that 'selection' carries connotations which, on Gould's reading, may not be warranted. Surely it's misleading to say that aspects of pure chance constitute "selective pressures" - since this phrase implies an ongoing circumstance that can steer the development of a species in certain directions.

Gould's point is that, while this does happen, a lot of the circumstances that determined which animals (and therefore genes) survived were not the result of such pressure at all, and thinking of all genes as having been "selected" is therefore misleading. If we use just "selection" as the term we run the risk of this *implying* "selective pressures" because (as you hint at) "selection" implies a directed process, not a chance process, or something in between.

Ultimately, if genes are there for reasons other than continuing selective pressures, then teleological stories concerning how the genes got there cannot be trusted.

Many thanks for your insightful comment!

[Bhengeveld]

Chris, thank you for your kind response.
To me personally the use of words like 'selfish' never implied teleological aspects, but coming from a teleological point of view before really reading about evolution perhaps made it clear to me from the get-go. (it actually hit me like a ton of bricks) All teleological points of view gone. But that's just my case, I can imagine people getting wrong ideas about it.

And I'm going to reread Mary Midgley :D

Btw: http://www.nytimes.com/2010/08/31/science/31social.html (in case you haven't already read it)
(considering your 4th myth)

[Chris]

Bhengeveld: thanks for the link! I had missed this story, and it amuses me to see this debate reopened. I can't help but chuckle when I see the ant used for research as if everything that works for insects works for all life. This was ever E.O. Wilson's problem. :)

All the best!