Read this and we’ll be done with Dobbs, unless he proffers another overhyped piece of science journalism.
Over at Richard Dawkins’s own site, he’s responded to Dobbs’s misguided critique of the “gene-centered” view of evolution as described in The Selfish Gene. Richard’s piece is called “Adversarial journalism and the selfish gene.” He’s remarkably polite for a man who’s been trashed in such an unfair (and erroneous) manner, and politely though firmly explains that, yes, he knows about regulatory genes and that, as we know, they’re simply selfish genes that regulate other selfish genes. He compares the toolbox of regulatory genes (a simile the biologist Sean Carroll also uses) to the subroutines of a Macintosh. and then notes:
Does Dobbs, then, really expect me to be surprised to learn from him that:
“This means that we are human, rather than wormlike, flylike, chickenlike, feline, bovine, or excessively simian, less because we carry different genes from those other species than because our cells read differently.”
Does Dobbs really think the existence of genes controlling the expression of other genes is either a surprise to me or remotely discomfiting to the theory of the selfish gene? Genes controlling other genes are exactly the kind of genes I have in mind when I speak of “selfish genes” as the “immortal replicators”, the “units of natural selection”.
Apparently Dobbs does think all this, and more. Dawkins patiently describes how he’s discussed this issue, the issue of interacting genes, the issue of environmental context in modifying gene expression, and so on, in his other books, books that Dobbs either read, forgot, or ignored. There’s a discussion of genetic accommodation, which to Richard (and me) also fails to kill off the selfish-gene idea. It’s a very temperate response, and very eloquent; I could evince neither trait had someone attacked me as gratuitiously as Dobbs did Richard.
Dawkins limits himself to just a tiny plaint at the beginning:
I have been asked to respond to an article by David Dobbs called ‘Die, selfish gene, die’[1]. It’s a fluent piece of writing featuring some interesting biological observations, but it’s fatally marred: infected by an all-too-common journalistic tendency, the adversarial urge to (presumably) boost circulation and harvest clicks by pretending to be controversial. You have a topic X, which you laudably want to pass on to your readers. But it’s not enough that X is interesting in its own right; you have to adversarialise it: yell that X is revolutionary, new, paradigm-shifting, dramatically overthrowing some Y.
. . . The Y in Dobbs’ article is my book, The Selfish Gene, and his main X is the important but far from new point that genes are not always expressed in the same way. He calls it phenotypic plasticity. Locusts are transformed grasshoppers: same genes, differently expressed. A caterpillar and the butterfly it morphs into have exactly the same genome, expressed in different ways. An animal is the way it is, not just because of the genes it possesses but because the context in which a gene sits affects how – and indeed whether – it is expressed. Dobbs makes some sensible points about all this, but there’s not a single one of them that I wouldn’t be happy to make myself – and in most cases did make, either in The Selfish Gene itself or in my other books. But his headline conclusion, namely that recent findings negate the thesis of The Selfish Gene, is not just untrue but deeply and perversely untrue.
and at the end:
But the fact that I didn’t go out of my way to stress [genetic accommodation] doesn’t even begin to mean that it is incompatible with the central thesis of The Selfish Gene. I can think of no reason why Dobbs should suggest such a thing, other than a journalistic desire to fabricate controversy where none exists. Which pretty much sums up his whole article.
I hope Dobbs learned a lesson from this, but I suspect he hasn’t, for he took a licking but keeps on ticking. The fact is that this kind of overblown journalism sells, and editors have no idea whether it’s scientifically accurate or not. “Evolutionary theory is wrong” is a trope I’ve heard my whole life, and though people like Dobbs, Massimo Pigliucci, Tom Nagel, Jerry Fodor, and others keep declaring neo-Darwinism dead, it refuses to lie down. Our only solace is that we have social media to call this nonsense out on the spot. I’ve done it, Richard has done it, and Steve Pinker has done it (see also Steve’s email quoted at the end of Richard’s piece):
I like “cuz,” which is probably there because of the Twi**er character limit but is still cute.
Subscribe.
sub
sub
sub
notify me, please, *%mm*t
I don’t know if you’ve taken a look at Dobbs’s response to your rebuttal, but on there he claims that he wasn’t attacking the “technical” explanation at all, but rather the metaphor that Dawkins coined to describe it. From the off, then, he’s downgraded his own article from ‘scientifically revolutionary’ to mere language pedantry, but even as that, I see no holes in Dawkins’s image.
“…he wasn’t attacking the “technical” explanation at all, but rather the metaphor that Dawkins coined…”
Gosh, no one’s ever done that before.
Yeah, I was thinking that, too. So Dobbs appears to have demoted himself to unoriginal pedant.
Yeah, that looks to me like balking. Someone who is known as a good writer and from what I read, is one, would not err so badly as to have so many people misunderstand.
Love Pinker’s sieve analogy!
Pinker is quoting some chap named Coyne in the last paragraph of this:- David Dobbs mucks up evoution part II 🙂
He was quoting Jerry!
Yeah! SOMEBODY didn’t read my post… 🙂
I loved that line!
Jerry, I think there’s a word missing:
“Dawkins patiently describes how he’s discussed this issue, the issue of interacting genes, the issue of environmental context in modifying gene expression, and so on, in his other books, books that Dobbs either didn’tread, forgot, or ignored.”
Also:
“I hope Dobbs learned a lesson from this, but I suspect he hasn’t, for he took
anda licking but keeps on ticking.”ehh my blockquote tags were screwed up, that last paragraph was mine the first Pinkers.
The concept of a gene isn’t directly tied to particular annotations of DNA sequence at all. In fact, the gene concept was created and matured well before the hereditary material was discovered. What genomics/molecular biology folks often call a gene (a transcriptional unit and its attendant regulatory sequence that often codes for a protein) is really just a shorthand for a class of sequences and doesn’t address all genes. Noncoding changes like changes in chromatin structure, GC content, repeat content, etc. all potentially underly genes identified through classical genetics methods (ie mutation screening, mapping, and complementation, etc).
There’s none so blind as those who refuse to see. Why do these people keep flogging a dead horse. I am not an educated person but even to me Darwin’s explanation is right
‘“Evolutionary theory is wrong” is a trope I’ve heard my whole life, and though people like Dobbs, Massimo Pigliucci, Tom Nagel, Jerry Fodor, and others keep declaring neo-Darwinism dead, it refuses to lie down.’
I heard Pigliucci speak a few years ago on the extended synthesis idea, and he expressly stated that the modern synthesis was correct as far as it went, but incomplete. He didn’t state that it was wrong. One possible source of confusion is that Pigliucci doesn’t use neo-Darwinism and the modern synthesis as synonyms, rather, he uses neo-Darwinism to refer to Darwin’s own revisions of his theory in subsequent editions of ‘The Origin of Species’ and other writings.
Dawkins is right that the perceived necessity of being revolutionary is driving such stories, but not just in journalism, it seems to me to be true in the scientific literature as well. Novelty uber alles is the current fashion.
I wouldn’t put Pigliucci in the same group with the other three. He is the only one of the four who knows evolutionary biology – the others not so much.
JAC, thanks for writing all these posts on this topic.
If you go back to the ‘one gene = one enzyme’ idea, and Crick’s ‘dogma’, I think you will see that the early focus on the gene was on the structural genes which code for proteins. That we have a term, ‘regulatory gene’ suggests that we discovered same and needed to differentiate it from the structural gene.
We are talking back into the 1950s or earlier here.
The term “gene” predates all knowledge of the structure of DNA, as well as the notion that DNA is even involved in heredity. It was meant to stand in for the actual cause of the phenotypic variability described in the recently (at the time) rediscovered work of Gregor Mendel.
I’m no biologist, but it seems to me Crick could say the same thing as Dawkins, that he wasn’t addressing structural genes specifically. The dogma describes how sequential information is handled through the modern hereditary system, see Fig. 1 of http://sandwalk.blogspot.se/2007/01/central-dogma-of-molecular-biology.html .
Moran: “Lewin defines the Central Dogma of Molecular Biology as,
“The central dogma states that information in nucleic acid can be perpetuated or transferred but the transfer of information into protein is irreversible. (B. Lewin, 2004)””.
So all genes, I think, are described under the dogma and with no specific focus as it is describing all observed pathways. (And, I think, that is also all pathways allowed under evolution.)
Oops. There are pathways observed later than the know ones when Crick stated his theory. (E.g. it was a good theory.)
I loved it when Richard Dawkins pointedly said this article was written to “harvest clicks by pretending to be controversial”. I just think the choice of the word “harvest” is perfect.
I also liked Steven Pinker’s “cuz”. Even if it’s a Twitter character limitation, it’s nice that a renown linguist like Pinker uses language as he does.
>We are talking back into the 1950s or earlier here.
Not early enough. The word gene dates to the turn of the last century (or even before), with de Vries using the word “pangen” at around 1890, Bateson using the word “genetics” in 1905-6, and Johannsen first using the word “gene” in 1909.
The earliest use of the word gene is best understood then as describing the unit of inheritance (at least, that’s how the founders of modern genetics used it), and terms like “structural gene” and “regulatory gene” only came much later during efforts to understand their biological function.
Whoops. This comment was supposed to be in reply to Jim above. Not sure how that happened.
Does that affect the argument though? Doesn’t gene-centrism still work? Perhaps Pinker meant that they’re trying to redefine it back to what it originally meant, but that what he, Dawkins etc are referring to are the modern notions of genetics and genes?
Oh and I forgot to say, that most likely Dobbs has learned his lesson even if he won’t admit to it. He probably wants to save face but won’t make this mistake again (I hope).
Alternatively, he may have learned that the best way to get a whole lot of attention is simply to strawman someone way more famous, then sit back and wait.
There seems to be a lot of this going around. Rational Wiki seems to be quite anti-Dawkins and anti-Hitchens. Check out their page on Dawkins (they also talk about refuting selfish genery and even have a separate article on it):http://rationalwiki.org/wiki/Richard_Dawkins. They take a stab at bashing Hitchens as well on his page. Interesting stuff, if misguided (in my opinion). Thoughts? I am nowhere near knowledgeable enough to talk about evolutionary biology, but surely someone here can discuss their points about the gene-centred view given on the Dawkins page?
Thanks for that link. The RationalWiki [RW] for Dawkins is very partisan without references to support the authors opinions. This whole bit below for example is entirely without any citations. I had no idea RW was so unprofessional. I hadn’t noticed this before because I’ve only ever used RW for looking up the mad & the bad. It seems I’ve let my bias lower my critical standards more than I’d realised. A good lesson for me…
It reads like lazy, hack journalism…
Yeah, that’s fairly sloppy. I didn’t even notice that before, I was more interested in the selfish gene part. There’s no evidence given that he’s alienated lots of people who fight unreason and who aren’t atheists. Obviously, he has done a lot for atheism and “conversion” to atheism and though most people might admittedly not be “New Atheists” there is still a lot of support out there. As for the whole idea he alienates non-atheists, er, that’s not quite true- he has religious friends and has even collaborated with religious leaders on the issue of creationism in UK schools- he wrote an open letter signed by both scientists and religious leaders to condemn creationism. And as for complaining about alienating NOMA-types (or anyone else for that matter): so what? He isn’t in the PR business, he doesn’t need to be liked by everyone. “Don’t take refuge in the false security of consensus” as Hitch would have said.
And again, as for the poor examples and so on, none of those are given and could probably be disputed. As for the Troubles, that simply isn’t true- in an essay in A Devil’s Chaplain, for example, he readily acknowledges that it has little to do with theology and everything to do with politics and economics- his point is religion acts as a label and so enhances the violence, not that it’s “just” religion.
A lengthy reply, and I could write so much more on that paragraph, but such nonsense deserves to be attacked.
Any thoughts on the gene-centred view dispute? Since that was my original point 😛
You are referring to THIS SECTION of the Dawkins’ RW. My only comment is that it’s poorly written ~ no laying out of the arguments. I’m with selection at the gene level [usually embodied in individual organisms] as opposed to group selection & Jerry has shot down group selection many times. Since I’m just a layman I recommend that you should START HERE where Jerry explains why [there are also three links early on in the post to other posts on the subject].
Are you the Canadian with a Youtube channel ~ involved in a lot of Youtube hangouts as a semi-retired philosopher? Uses similar name to yours.
Thanks for the links- I’ll be sure to check them out. I’m also a layman, so don’t expect me to grapple with the technical stuff. But the stuff on this section of the wiki seems to be mere assertion or stuff that’s been rebutted. I’m not sure what they mean by more broadly defining inheritance though (“ecological inheritance”).
And no, I’m not the Canadian, alas.
http://en.wikipedia.org/wiki/Niche_construction
Actually, niche construction and inheritance right there. Google always has the answers!
I’m thinking that “ecological inheritance” has something to do with the Extended Phenotype [EP] which is Dawkins’ principle contribution to evolutionary theory [he says so] & he’s written a fine book going by that name. I suppose it MIGHT refer to say dam building in beavers, but that’s a wild guess as the term is new to me & I suspect it’s not a term that Dawkins would approve of if it is suggesting other than a gene-centric view…
The EP idea is gene-centric
didn’t notice your new comment about niche construction before posting my EP comment.
Yeah, reading the niche construction description and the ecological inheritance page doesn’t seem to me to be too revolutionary. I don’t see how it contradicts genes leading, really. Perhaps I’m missing something but to me selfish gene theory and niche constructionism/ecological inheritance are compatible.
Nothing more than “extended phenotype” (i.e., ecology) with co-evoution (i.e., classical Darwin). Just a made-up name for common knowledge.
BUY IT
Run out of wine so up the stairs to Bedfordshire
I’m definitely not the Canadian. Now that I’ve seen him I know for sure :P.
For the record, I’m Scottish, so if a Scottish Ozymandias pops up it probably will be me!
W. Benson, yeah I’d agree. It seems like they want to attack him for the sake of attacking him without actually attacking him…it says ecological inheritance differs from genetic inheritance in that it means the descendants have to continue modifying the environment (http://en.wikipedia.org/wiki/Ecological_inheritance) but I don’t see how the notion affects gene centric ideas or the centrality of genetic inheritance. I’m no expert, I admit, but I can’t see what this contributes.
The Canadian:- Ozymandias Ramses II
“philosopher of atheism”. Sigh.
As Coel notes in a post on his bl… website, agnosticism isn’t the same as atheism – you can be an agnostic theist – and I claim that is the better of two bad candidates for philosophy.
Atheism is in all its forms an observation on a religious existence claim (existence and so actions of purported gods), and knowledge on what exists has been solely acquired by empirical methods.
Agnosticism is in all its form an observation on some knowledge claim. Arguably the idea that it is an area of philosophy is slowly dying, e.g. see Jerry’s discussions on science and its knowledge.
Frankly, since theology is the only defense for NOMA, the implied but erroneous claim that atheists need to use philosophy becomes the accommodationists last Catch-22 holdout to keep NOMA from analysis.
If these Dobbs-refuting articles are examples of what your next book is going to be like, I’m looking forward to it already. Clear, concise and with a nice sharp edge. Meanwhile, Dobbs has backed up so far into ‘I just don’t like the metaphor’ territory he might as well have not written the damn article in the first place.
Shirley, I can’t be the first to comment on the fact that Professor Ceiling Cat is also Richard Dawkins’s goto guru on population genetics?
(Of course, goto statements are generally considered harmful, but we’ll let that slide in Richard’s case since programming was always a side gig for him.)
Cheers,
b&
Yes, not goto. Properly structured procedure call. Good call Ben!
Bah, a subroutine call and a return statement are just goto equivalents. So are for/next loops and if/then/else.
Goto isn’t inherently bad, it just lends itself to undisciplined code. A disciplined coder can write beautiful code using goto. I’ve written some lovely assembler language programs in days gone by, and I’ve seen horrible spaghetti code written in goto-less high-level programming languages.
Oh, of course — a great craftsman is at most annoyed by “bad” or “worng” or “less-than-ideal” tools, and will often take perverse delight in doing that which supposedly can’t be done with a particular toolset, and doing it well.
And, to be sure, assembly is its own hairy beast that bears at most logical equivalence to today’s highly-abstracted languages, especially for the niche cases where it’s still (appropriately) used.
But, aside from those types of exceptions, in the everyday life of the boring real world…GOTO is almost always a sign of an inexperienced (or, at best, a lazy and / or hurried) programmer who hasn’t properly mentally abstracted the business logic of the problem being solved.
Any remaining exceptions, where it’s actually properly used, are likely to be programmatic expressions of real-world business logic that is, itself, highly problematic.
That’s actually one of my biggest professional frustrations: having to write insane code not because I’m personally insane, but because the request itself is insane. Have you ever written code to calculate the means of sets of numbers that are themselves median values of aggregate figures? I have….
Cheers,
b&
“Have you ever written code to calculate the means of sets of numbers that are themselves median values of aggregate figures? I have…. ”
I have no idea what the mathematical significance of that result would be and it may be that the person who asked you to do that didn’t know either.
I’m pretty quick to tell the business side that their requirements are incoherent. They seem to think that developers can write programs or reports that miraculously solve problems that can’t even be defined.
But you’re right that a Goto can be a warning flag; I don’t think that I’ve used one since leaving Fortran. I reserve the right to, though. 🙂
Oh, there’s no “maybe” about it….
We used to, for years. We still do for the more egregious examples…but more for the comedic effect than in the expectation that it’ll actually do any good.
That I think we could handle. What tends to throw us for a loop is that they want us to solve problems that nobody’s even thought of yet….
After all that ranting on my part…I have to admit that, until recently, we actually did use GOTOs a fair amount. You see, before the latest incarnation of Microsoft’s SQL server, it was the only way to do error handling in a stored procedure….
b&
Ah, yes, wasn’t thinking of that; I guess I have used a Goto in that context. These days, I tend to only use SPs to return data to business objects, and let the application code handle any errors.
We’ve got lots of standalone stored procedures that we use for automated tasks (run as jobs by the server itself) — things like shipping data around, validating setup information, running reports and emailing the results, that sort of thing. That’s where you’ll find the error handling.
Regular reports / Web applications / whatever we generally don’t handle and just let the error bubble up to whatever level that we’re interested in doing something about it.
b&
….and when you think of your cron jobs remember that cron should be spelled chron if it refers to time (chronos); if follows cron refers to Cronos – the dude in Hades that is tormented for eating his kids.
I even made a meme of this once.
Either memory was way more expensive than I can fully comprehend in the late ’60s and early ’70s; or Thompson, Ritchie, and Kernighan had really bad keyboards; or they simply couldn’t spell worth a damn.
You see that sort of thing everywhere, even in situations where economy-of-typing doesn’t apply.
And then there’re the peculiarities in the other direction, like adding the “a” back in “daemon”….
b&
They probably add the “a” back because when you say, “a demon” is running something, people get nervous. 😀
You’re right.
They should have called them, “Jesuses,” and submitting a cron job to be a “prayer.” Then, you wouldn’t kill the job; you’d crucify it. And you wouldn’t nice a process, but damn it for positive (low) priorities and beatify it for negative (high) priorities.
Somebody could have fun with something like that….
b&
That’s a common usage, and I used to do that, but I started writing those as console jobs in C#, and using Task Scheduler to run the jobs. C# makes much prettier code than T-SQL, and funnels all functionality through our business objects, which handle data validation, logging, auditing, error handling, etc.
I have all the fire of a TV evangelist about this being the absolute best way to do things. 🙂 My business card reads “Have Framework, will Travel.”
“Business objects” — see, that’s just it. We’ve tried multiple times to create some kind of a coherent framework…but everything we do is its own special snowflake. Over here, this class of employees is defined as those with this set of job codes; over there, the same set is defined as a similar-but-slightly-different range of job codes. This sales category is the same for all stores except for these three. And so on.
It all superficially looks like you should be able to construct some frameworks of whatever variety to abstract it all out…but then you get nibbled to death in the details. Pretty soon, each instance needs more than enough customization to offset even any lines-of-code savings you might have originally had. And then they go and change this one report / whatever, and only that one, and now you’ve either got to re-write that from scratch with a copy / paste duplicate of your framework (or whatever) or you’ve got to beak all the others or write insane logic to undo and redo math or…
…or, you just give up and treat everything as its own little snowflake, which is what the business model actually really is, even if nobody ever wants to admit as much.
TL/DR: nothing we do repeats itself, even though most things rhyme. Maybe. Sometimes. For a while, at least….
b&
“It all superficially looks like you should be able to construct some frameworks of whatever variety to abstract it all out…but then you get nibbled to death in the details.”
Yep, I understand. I was able to create the framework because for a brief, shining, three-year moment, I was in charge of the IT department and I was able to get the business requirements altered to my liking. 🙂 Small company, too, so no other departments to fight.
IT really needs to be involved with the design of the business requirements, because we understand how a disciplined, organized approach to solving business problems translates to a more elegant system design, which benefits the business.
Most companies, though, view us as factory workers who should only do as we’re told.
Factory workers — ha! Such respect!
We’re somewhere between plumbers and janitors….
But, yeah…the organization and consistency we’d be able to provide to the business models would benefit a lot more than just the code we have to write and maintain.
b&
Hey… “Goto” can be extremely useful, especially when instructing an idiots where direct himself.
Yes, I guess a subroutine call doesn’t have the same effect, since it implies a return.
You know…you raise a good point. That probably explains the Jesus incident: somebody told him to gosub Hell, rather than goto, and forgot to check for a return value. Gives new meaning to the term, “zombie process!”
b&
Chucks, I don’t know whether to goto that link or not. (^_^)
I suspect many press articles about Richard Dawkins are less interested in holding water, than boiling it.
A short one to note that Dobbs revised his original article significantly. He also explained why and documented the revisions here.
I still don’t agree with all he says but appreciate the effort, I’ve commented this whole debate here.
I am genuinely surprised by his latest effort: he corrected a lot of the science and showed a remarkable level of integrity. I just wanted to put it on the record, maybe Jerry will want to follow suit.
Correction of the last link (apologies) the correct url is:
http://wp.me/p3NcXb-41
It’s still very misleading, since he’s referring to adaptive traits originating “through gene expression alone”. There is no reason to believe (and lots of reasons to disbelieve) that complex adaptations like increased muscle development in response to exercise, or the grasshopper’s locust phenotype, originated without any genetic changes.
That’s agreed, no reason at all to think traits don’t have a genetic origin.
That’s why I’ve said:
On the other hand, Dobbs’ take-two added lots of clarifications, both on the intended main effect (popularise a level of complexity that is usually ignored) and on plenty of scientific details. He did try to take-in and account for all the (sometimes hard, even confrontational) criticism. Very few people ever do that, most just entrench and play the victim. So I felt it was important to point out the shaky science (still) but also paramount to acknowledge the effort and intention.