In defense of the Selfish Gene

This is a quick rebuttal of the thesis explained by David Dobbs in his recent “Die, selfish gene, die” Aeon Magazine article [Edit: 14/12/2013 now archived here]. It is a well written, well documented and quite long piece that tries to explain why we should consider the “selfish gene” metaphor outdated and substitute it with more “correct” ideas. The article shows two, very rare, and highly appreciated qualities: the author is very careful in pointing to all his sources (he had plenty) and also did manage to get Richard Dawkin’s opinion directly. I want to point this out because both things are rare, and essential for good science communication. The rest of this post will be highly critical of the article contents, so I would like to make sure the reader remembers this short initial praise: it is deserved and nothing of what I’ll say below invalidates it.

Having said this, the content is so wrong that I actually had to force myself to read it all, it was intellectually painful, but I’ve managed.
Dobbs cites Gregory Wray, a biologist at Duke University in North Carolina:

‘Different groups of animals succeed for different reasons,’ says Wray. ‘Primates, including humans, have succeeded because they’re especially flexible. You could even say flexibility is the essence of being a primate.’

and then adds:

According to Wray, West-Eberhard and many others, this recognition of gene expression’s power requires that we rethink how we view genes and evolution. For a century, the primary account of evolution has emphasised the gene’s role as architect: a gene creates a trait that either proves advantageous or not, and is thus selected for, changing a species for the better, or not. Thus, a genetic blueprint creates traits and drives evolution. […] But a number of biologists argue that we need to replace this gene-centric view with one that more heavily emphasises the role of gene expression — that we need to see the gene less as an architect and more as a member of a collaborative remodelling and maintenance crew.

I can live with all this, as long as one notices that the ability to dramatically change the gene expression patterns is a ‘trait’ in its own right, that this ability is inherited like any other trait, and that (unless a different carrier of inheritable information is identified) the information that makes this trait possible is stored in the subject’s DNA (somewhere, somehow, the details are irrelevant for now).

Dobbs also cites Michael Eisen, an evolutionary biologist who researches fruit flies at the University of California, Berkeley:

‘It’s not that genes don’t sometimes drive evolutionary change. It’s that this mutational model — a gene changes, therefore the organism changes — is just one way to get the job done. Other ways may actually do more.’

and then adds:

Like what other ways?
There are several, but one called genetic accommodation is, according to West-Eberhard, particularly powerful and overlooked.

West-Eberhard is Mary Jane West-Eberhard, a wasp researcher at the Smithsonian Tropical Research Institute in Costa Rica who appears to have inspired most of the article. In fact, this is where the whole argument derails beyond repair.

The first warning sign:

[genetic accommodation] takes a moment to explain. But bear with me a moment, and you’ll understand how you, dear reader, could evolve into a fast and deadly predator.

What? I can personally evolve into something else? I can change my own genetic blueprint, in a meaningful way during my life (leaving alone notable exceptions such as antibody recombination)? Seriously? Wow, I’d better read on! (In fact, by this point I was already forcing myself to keep reading)

For example, suppose you’re a predator. You live with others of your ilk in dense forest. Your kind hunts by stealth: you hide among trees, then jump out and snag your meat. You needn’t be fast, just quick and sneaky.

Then a big event — maybe a forest fire, or a plague that kills all your normal prey — forces you into a new environment. This new place is more open, which nixes your jump-and-grab tactic, but it contains plump, juicy animals, the slowest of which you can outrun if you sprint hard. You start running down these critters. As you do, certain genes ramp up expression to build more muscle and fire the muscles more quickly. You get faster. You’re becoming a different animal. You mate with another fast hunter, and your kids, hunting with you from early on, soon run faster than you ever did. Via gene expression, they develop leaner torsos and more muscular, powerful legs. By the time your grandchildren show up, they seem almost like different animals: stronger legs, leaner torsos, and they run way faster than you ever did. And all this has happened without taking on any new genes.

Then a mutation occurs in one grandkid. This mutation happens to create stronger, faster muscle fibres. This grandchild of yours can naturally and easily run faster than her fastest siblings and cousins. She flies. Her children inherit the gene, and because their speed wows their mating prospects, they mate early and often, and bear lots of kids. Through the generations, this sprinter’s gene thus spreads through the population.

Now the thing is complete. Your descendants have a new gene that helps secure the adaptive trait you originally developed through gene expression alone. But the new gene didn’t create the new trait. It just made it easier to keep a trait that a change in the environment made valuable. The gene didn’t drive the train; it merely hopped aboard.

Right, I’d better study more English, ’cause I thought that fast and quick were synonyms, but I’ll leave semantics aside and dive into the core argument.
The citation above is the key argument here. We should remove genes from the centre stage of our understanding of evolution and replace it with something else, presumably the environment-dependent variability of gene expression, or maybe the concept of genetic accommodation itself. We may do that, but it is crystal clear that the variability of gene expression is the result of some more genes, their interaction with non-encoding parts of the DNA sequence (those little things that we molecular biologists call promoters) and all the associated machinery.
What this means is that the variability of genetic expression is a genetic trait, making the story above incomplete: it lacks a crucial detail about what happened before. The genetic-expression variability that allows our predator to “train” and get faster exists as the direct result of natural selection itself, and the trait that was positively selected is the ability to change the expression patter so to build more muscle. We know this because genetic expression is demonstrably regulated by proteins that interact with specific DNA sequences (as well as other systems that act further down across the line, changing what happens to messenger RNA and so forth). In other words, it all boils down to other genes (the ones that encode for the regulators) or DNA sequences (the promoters and similar trickery). So no, we don’t need to put aside the “selfish gene” idea, we may acknowledge that things get fantastically complex in almost all cases (very rarely one gene encodes for one and only one phenotype), that most biologically interesting phenomena happen because of the interplay of genetic networks (where one up-regulates the expression of some, down-regulates some others, and all genes involved may have some other effect elsewhere), and that saying “one trait corresponds to one gene” is usually wrong.

But that is all. Claiming that we need new evolutionary concepts to replace the faulty idea of the selfish gene is just downright wrong. What we do need to understand and model in new and more reliable ways is how a genome entirely made up of selfish genes (and other selfish non-coding sequences) can evolve to create monstrously complex networks that allow the astonishing adaptability of human bodies. The challenge is to see how, why and when “selfish”* elements associate and “collaborate” in ever more complex ways (generating the variability that is impossible to pin down to a single gene). I will not discuss this here: there are lots of things we understand about this mechanism, and lots more that are being researched intensely, but the challenge is about how “selfish” elements invariably produce “collaborative” networks in which each single element depends on all the others. This is important, and very interesting, because it happens on many levels: for example, in a previous post, I have hinted about this kind of problem in the domain of human collaboration (see the “Edge” link).

But for now, I’ll finish it here: Dobbs claims, with the crucial support of West-Eberhard that:

By the time you’ve finished his book, or well before that, Dawkins has made of the tiny gene — this replicator, this strip of chemicals little more than an abstraction — a huge, relentlessly turning gearwheel of steel, its teeth driving smaller cogs to make all of life happen. It’s a gorgeous argument. Along with its beauty and other advantageous traits, it is amenable to maths and, at its core, wonderfully simple.

Unfortunately, say Wray, West-Eberhard and others, it’s wrong.

And that’s pure nonsense, bad reasoning and misinformation.

*Note the quotation marks: “selfish” and “collaborate” are used here as metaphors. There are very little reasons to believe that genes are conscious, so they can’t be “selfish” or “collaborative” as humans would.

PS: (added on 14/12/2013) David Dobbs has published a new version of his article, I’ve responded to it here.

Tagged with: , , , ,
Posted in Evolution, Science
15 comments on “In defense of the Selfish Gene
  1. Michael Fugate says:

    Nicely done.

  2. Paul LaMar says:

    Dobbs’ description of The Selfish Gene (TSG) left me wondering if he had read the book in its entirety. Dobbs seems to create a “straw man” selfish gene theory in order to provide shortcomings that are not actually in Dawkins’ writing. (Perhaps he was trying to point out that the public understanding of the role of genes and evolution is flawed. Still, the failure of the public to understand TSG in no way implies TSG is “wrong”.)

    Thank you for taking the time to post this. You and Jerry Coyne have done an excellent job pointing out some of the article’s most serious errors. It’s this kind of critical and thoughtful feedback that helps an amateur enthusiast like myself avoid muddled thinking about evolution.

    One nice thing about all of this is I’m learning which evolutionary bloggers and journalists understand evolution and which don’t.

    I first read TSG in the 1980’s and it helped me avoid wasting time reading about the various flawed “group selection” theories that were in the popular press then. Now, more than thirty years later, TSG is performing the same sort of function with the flawed “genetic accommodation” theories gaining traction in the press today. TSG is truly a science classic.

    • Sergio Graziosi says:

      I don’t know what to say, you made me blush. Thanks. As I’ve said on Jerry’s blog I’m genuinely surprised by how Dobbs article has been accepted by plenty of scientists that I consider top-class in their own field.

  3. […] I tweeted to Jerry Coyne, asking him to engage in the debate; but I couldn’t wait, so I wrote my own rebuttal as well. Both my actions were far from friendly: my own piece is pretty harsh, and Jerry Coyne […]

  4. […] impulsive reaction to a widely inaccurate account of the Selfish Gene idea published on Aeon. The first post was inspired by my strong reaction: I was convinced that the Aeon article was “a mean (and […]

  5. […] they self-sustain. It is exactly the same mechanism that can be seen in biological evolution: genes can be seen as selfish, because they will survive multiple generations if/when they facilitate their own propagation; […]

  6. […] the resulting posts also happen to be among the most popular on this blog, see for example here (with follow-up) and […]

  7. I have not read Dobbs’ article, beyond the quotes your provided. So the rest of my comment is based on those quotes and all the good arguments I know against the selfish gene. I disagree with you quite vocally below. The hope is that you will respond equally vocally and we will both grow from the interaction. I apologize for any hostility in my text, it is not intended.

    First, an important thing to get out of the way: nobody is disputing evolution, nobody is disputing that evolution is grounded in the physical. What is under dispute is what kind of metaphor is most useful for thought and easiest to use to see the current and future directions of biology. For this, you have to outline the boundaries of where the “selfish gene” ends, what constraints it places on you in shaping your thought. What questions does it suggest? Where does it draw the focus? In your defense the selfish gene, it seems like you were defending a statement almost as basic as: “Protons, electrons, etc ground biology, and those aren’t wrong”. Nobody has disputed that. What they have disputed is where thinking about genes points you.

    The whole point of the extended evolutionary synthesis and other modern trends in ecology and evolutionary biology is to realize that genes (implemented in DNA or otherwise) is only one of the many ways that hereditary information is transmitted between (or within) generations. Other ways include epigenetic proteins, shared environments, and parent-to-offspring teaching (or other more primitive forms of ‘booting up’ that exist at all levels, from the single cell up). In many cases, these other factors are much more important than DNA. Sure, for some of them you can add epicycles upon epicycles on the ‘selfish gene’ and throw around empty words like ‘complex networks’ and such, but at some point, it becomes more productive to operationalize your theory at a different level. That is what biologists are doing now, and Dobbs (at least from the examples you provide) seems to capture that spirit relatively well. I really like that predator analogy. In particular, the lock-in aspects at the end is a nice highlight of the Baldwin effect.

    Sorry if this comment is coming off as a little hostile. It is just that the selfish gene metaphor, especially when it is combined with the more nefarious force of cheap genomic sequencing, is a huge roadblock in many disciplines. Especially in cancer where people still keep searching for gene mutations, while they should be chronicling disturbances in the microenvironment. In fact, for any evolutionary process within a single individual — where pretty much all cells are nearly genetically identical — the selfish gene view becomes convoluted. It is always the developmental history and local environment of the cell that is the central player (that is why we still classify cancers by tissues of origin). Sure, you could bend-over-backwards and say “well, they just misunderstood the selfish gene, you should always consider how the gene interacts with each particular environment”. But at that point, why prefer the ‘selfish gene’ to the environment? Especially since the environment is such a central player.

    A commentator in this thread accused Dobbs of straw-man-ing the selfish gene, while Dobbs seems to be actually be channeling the spirit of the last couple of decades of biological research relatively well. Instead, your response — like many selfish-gene evangelists I have read on the internet — seems disconnected from and unaware of current challenges in biology.

  8. Sergio Graziosi says:

    Just a quick note to say that I’m writing a proper answer, but work takes precedence :-\.
    No worries and no hostility detected: challenges to the arguments I make are more than welcome!

  9. […] pessimist, especially one that argues against the gene-centric view of evolution, can suggest that genetic information is just one of the many heritable channels. Non-genetic […]

  10. Sergio Graziosi says:

    I hope you’ll find it very hard to hurt my feelings by strongly objecting to my arguments, I’m sure it’s possible, but hopefully not easy. In this case, the only criticism directed to the person and not the argument is entirely right, so not a reason to perceive it as hostile.
    [CharlesInVT’s blog is indeed packed with very good stuff! Thanks for flagging it up, would have missed it otherwise.]
    I’m afraid I won’t be able to provide strong counter arguments, because if there is any disagreement it is on tangential details.
    Where do we disagree? I think it boils down to how we interpret Dobbs:

    In your defense the selfish gene, it seems like you were defending a statement almost as basic as: “Protons, electrons, etc ground biology, and those aren’t wrong”. Nobody has disputed that.

    You are right that my defence was of the very basic concept. At the time I did feel the basic idea was the intended target, unfortunately the original version of the article is now gone (the one that is still on Aeon is the much improved version 2), so I can’t check how much my impression was justified: we’ll have to rely on the quotes I’ve included at the time.
    This leads to the only remark you make about me the person:

    your response — like many selfish-gene evangelists I have read on the internet — seems disconnected from and unaware of current challenges in biology

    This was entirely correct (if directed to the version of me which existed back then) and is still mostly correct: I did catch up a little since writing this post, but I’m still professionally engaged on other stuff, so I am disconnected. I do try to keep in touch, but for my personal pleasure, so I’m always lagging behind a fair bit. At the time, the early days of this blog, I was just starting to look around, and this particular debate did give me a valuable incentive. I do wonder how I would have read Dobbs’ article if I stumbled on it today for the first time, perhaps differently, perhaps not.

    To put it bluntly: I think the only disagreement between you and the 2013 version of myself is that the latter took “Die, selfish gene die” and “Unfortunately, say Wray, West-Eberhard and others, [the selfish gene idea is] wrong” at face value, while you, knowing beforehand the arguments from the inside, read it in a much milder sense. My reply here was simply: “no, the selfish gene idea is not wrong”. I can still stand behind this point, but I can also qualify in ways that show why I don’t disagree with the spirit of your own comment.

    Basically, what you are saying is that in an ever increasing number of fields, “it becomes more productive to operationalize your theory at a different level” i.e. leave the selfish gene metaphor behind (or in the background) and try to produce new theoretical instruments that work better for a given perspective.
    To this, I agree with enthusiasm, so we probably won’t have much to learn from arguing (I seriously doubt I have much to teach you!), the only point we might contend is the following:
    for me, one could and should look for “new theoretical instruments that work better for a given perspective” without feeling the need to say that whatever works from a different angle “is wrong”. Naturally, “Not fit for the current purpose” is fine. But two theories that have overlapping subjects can be different, hard to unify, and still both useful in their own way (each with their own different focus). When, like me, you put little ontological weight on theories and are happy to judge them instrumentally (how good are they to produce reliable predictions in their own domain), there is no tension left between extended synthesis and selfish genes, or between group selection and selfish genes, for that matter.
    It’s remarkable how we tend to gravitate around repeating “themes”, here the thing that springs to mind is a conversation we’ve had around my automatic aspiration to merge, unify or reconcile seemingly irreconcilable theoretical stances. The reason why I did “throw around empty words like ‘complex networks’ and such” is because of this, striving towards reconciliation is an urge that I have, one over which I have little control. Given the fact that I’ve been and still am outside the field (just a keenly interested spectator), vague handwaving was the only thing I could do!
    Thus, for the main point you’re making:

    In many cases, these other factors are much more important than DNA.

    I can’t propose any vocal counterargument, because I agree with you. Your hints at carcinogenesis are (unsurprisingly) particularly relevant [Side note. You brought back a strangely fond memory of my awkward feeling when discussing oncogenes during an undergrad exam in the previous century: at the time I couldn’t quite pinpoint where my doubts were coming from, but I still couldn’t fully commit to the “genes only” view I was expected to explain, leading to an interesting and deeply uneasy (for me!) discussion during the exam].

    Anyway, my main point here is that I’m allergic to hasty simplifications. In this case, one can and should realise that in many fields the “selfish gene” metaphor may not be the best starting point to produce useful theories. This does not mean that the “selfish gene” metaphor is “wrong” and should be killed. It means that people should pick and develop the right tools (shades of our old discussion here!), whenever they need them, and that the drive towards conformity, which pushes towards always using the same egg (or few eggs) from the same theoretical basket is deleterious.
    I’m sort of confabulating on why I got angry: saying the SG idea is “wrong” means that one accepts the wrong background trend, a silly form of naive positivism (theories are either right or wrong) and makes it harder for the next breakthrough to happen.
    Instead, realising that a given theoretical approach isn’t fit for a given purpose, and that trying to develop new approaches is probably better, should not come with the additional burden of demonstrating that “theory X is wrong”. Once the new theory Y is well formed, people will face the additional intellectual challenge of trying to bridge/merge/reconcile X and Y, which happens to be the kind of activity that excites me (and perhaps you as well).

    In Science Communication terms, which is what Dobbs was doing, I find it irritating when people try to stir public opinion away from one over-simplification (e.g. 1 gene = 1 trait) and do so by offering and equally wrong one instead (the SG idea is “wrong”). You can find traces of this knee jerk reaction of mine all over the blog, including the latest one on Brain and computations. Same stuff: if someone tries to correct an error with another error, my blood pressure rises…

    Right, it took me many days to write the above, and even so, it still looks to me like a disorganised stream of (little) consciousness. Hope you can parse it nevertheless.

    • Sorry for the slow response. WordPress doesn’t notice me if you respond as a top level comment.

      I wanted to pick up on one thread from your response. You wrote that the metaphor isn’t “wrong” and instead we should embrace plurality and instrumentalism, and avoid the positivistic distinction between “right” and “wrong”. Here we might have been talking on crossed lines. Note that I did not use “false”, but “wrong”. That means we have to ask “wrong for what?” (Although similar things should probably be asked of false, too, but that is a more general discussion).

      When I said the selfish gene is wrong or bad, I meant that it is wrong or bad for the development of biology in the near future. As a metaphor, the mindset it puts us in holds us back from the sort of questions we should be asking and avenues we should be exploring. This doesn’t mean that nobody should think (or do research) framed by the selfish gene metaphor, some minimal level of plurality is always good. But if one wants to understand why the current (and near future) literature is going in the directions it is going,or why scientists are asking the questions they are asking then thinking of biology in terms of the selfish gene will make that more difficult to make sense of. In that sense, it is wrong.

      One of the goals of the science communicators that communicates current events (not sure why this is so popular, but it is) is to give people the tools to think about current science. The selfish gene is not such a tool, and in fact inhibits this. So I think it is perfectly fine for a science journalist to say “die selfish gene, die”.

      When I am talking to a doctor, for example, I don’t want to have to always jump through the epicycles of explaining something in terms of the selfish gene. At some point, I just wish that the doctor would take a little bit of time to retool their cognitive toolkit with a more current metaphor for biology. Unfortunately, one as simple and catchy as the selfish gene doesn’t seem to exist. Hopefully, some science journalist — the next Dawkins — finds one soon.

  11. Sergio Graziosi says:

    my turn to apologise for being slow. One reason is that I had nothing vaguely interesting to reply. The wishes you express in the last paragraph are agreed: the cognitive attraction that the selfish gene metaphor is exerting over people who should know better is probably too strong, and I certainly am somewhat caught in its area of influence (also agreed: using “wrong” instead of “false”, implying that the question “wrong for what?” is fundamental). Moreover, indeed it would be best if we had a new, equally “attractive” metaphor to latch on, one which happened to be better suited to tackle the present challenges (multifactorial traits and diseases, ecological disruption, cancer, of course, and so on).

    I’m writing this comment now because, struggling as I am to come to terms with the outcome of the self-destructive UK referendum, an old worry of mine is re-surfacing in all its ugliness. What if we are reaching a new plateau in scientific growth, where “progress” (whatever that is) is slowing down considerably? I’m worrying that this might be the case because we might be reaching a hard to break cognitive limit.

    I’ll use physics as an example because it’s easier to explain. Quantum theory is notoriously hard to master: one needs something like 2-3 years of study/practice to master the mathematical tools required to get started, on top of that, probably the same amount of time is required to learn the ins and outs of the theory itself. At this point one may be able to have a good understanding of the current challenges, say the struggle to combine QT with relativity/gravitation, or the quest for refining our understanding of fundamental particles/waves/whatever. In all this, the philosophical implications loom and it’s notoriously hard to produce “interpretations” of QT that happen to make intuitive sense (e.g. conform to our natural intuitions of how the world works).
    My worry: a small proportion of human beings come equipped with the qualities required to easily master the above, it can be mastered, but for almost everyone it’s hard and requires dedication. Once mastered, it’s likely that you’ll be so immersed in your theory of choice that it’s going to be exceedingly hard to think out of the box and come up with a new imaginative way to propose new theoretical outlooks. Thus, the probability of getting a new Einstein may be dropping at an alarming rate. Moreover, in the mean time, the inertia of the current theoretical approaches is increased: on the personal level it was very costly to master them, so people will be less willing to let them go (see my own attachment to the SG idea). Sociologically, the way science currently works means that new “disruptive” theoretical approaches face a range of “getting started” barriers that has never been stronger.
    Thus, all this factors effectively make it harder to both produce and then disseminate new promising (but still comparatively underdeveloped) ideas.

    If we go back to biology, something similar seems to be happening: drifting away from a gene-centric view is hard, requires to master the mathematical tools you know so well, and then to depart for new uncharted territories. There is a fair chance that there is no new metaphor on offer: perhaps the SG is a local optimum in cognitive space, it’s simple enough to be quickly understood and allows almost everyone to see its explanatory power. Maybe, just maybe, the new theoretical developments are drifting away from the cognitive sweet spot, they become too complicated, they require too much investment to master, so they are only available to specialists. Once this happens, specialist outlooks get locked-in, and the associated ideas start getting siloed in their respective sub-cultures.
    If I’m right, this means that new theoretical breakthrough are becoming harder, much harder. It’s one reason why I’m so happy you’ve found me: you are clearly one who is trying to push against the current constraints – your urge to jump across specialist silos is all too evident… I don’t think I can help you much, but I can at least show appreciation and encouragement, FWIW…

    Coming back to the subject: I do see that you are trying to say that remaining locked into the SG metaphor is actually making a lot of stuff harder to grasp. You’re right, of course, and I assume we both would welcome a new metaphor that is equally catchy but fits better with the current challenges. My worry is that there isn’t one (not equally catchy), but I’d love to be proved wrong (of course, I would love even more to prove myself wrong!). Because of my inclination I’d prefer to find something that is easy to reconcile with the SG idea, so my own search space is limited, and the probability that it contains what I’m looking for is even smaller.
    As I’ve said at the beginning: it’s hard to find where we fundamentally disagree (but if keep spilling my beans I’m sure you’ll find something ;-)).

Leave a Reply to Systemic change, effective altruism and philanthropy | Theory, Evolution, and Games Group Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Follow me on Twitter

All original content published on this blog is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Creative Commons Licence
Please feel free to re-use and adapt. I would appreciate if you'll let me know about any reuse, you may do so via twitter or the comments section. Thanks!

%d bloggers like this: