[Updated]
Warning: Biology wonkage, proceed at own risk
On the forum at Richard Dawkins’s site, I came across an exchange between Dawkins and a forum member describing “natural selection” as the “driving force of evolution”.
I’m bothered by this usage. For one thing, it seems like the kind of thing that can contribute to the rancorous debates in biology that sometimes strike me as more semantic than scientific. However, it is also possible that I’m not quite understanding what the intent of the phrasing is.
The discussion at Dawkins’s site was started by “Hyrax”, who was puzzled by the assertion that evo-devo poses a challenge to the theory of natural selection. In a question to Dawkins, he/she/it/not-rodent writes:
I have been involved in a lot of debate over the subject of ‘Evo Devo’ and similar ideas that claim to usurp natural selection as the principle driving force of evolution.
From Dawkins’s response:
Of course evo-devo doesn’t usurp natural selection, any more than a car’s engine usurps the driver. There are alternatives to natural selection, and the obvious one is random genetic drift. In its molecular guise of Kimura’s ‘neutral theory’ it is plausible as an alternative driving force of evolution. Indeed, I think it is quite likely that the majority of evolutionary change at the molecular level is neutral. But natural selection is the only known evolutionary driving force that is capable of generating adaptation.
Perhaps this is just a matter of definitions, but, by my understanding, natural selection is an utterly passive process; it can’t “drive” anything. Evolution requires variation — genetic mutation, recombination, epigenetic twiddling. The possibilities manifested out of that variation are then filtered based on environmental pressures (broadly conceived). That filtering process is what is called “natural selection” — it does not “drive” evolution any more than a vacuum cleaner filter is a source of suction.
Note what Dawkins is saying: natural selection generates adaptation. AFAIK, natural selection can not “generate” anything at all — it can only, well, select. An “adaptation” is just some bits of variation that have survived and can be retrospectively crowned as more useful — for the situation under consideration — than the available alternatives that didn’t survive.
So I’m curious: why does this metaphor stay around? Its convenience as a turn of phrase overcomes the degree of deficiency it bears from being misleading? Dawkins seems to use it quite unselfconsciously, and I know I’ve seen it in other places. Does anyone else find it nettlesome, as I do, or am I being hyper picky?
I have wondered whether it arises from something in the mathematics used to describe “fitness landscapes”, i.e. that something about the mathematics somehow makes the metaphor work when applied in that context. It also seems possible that there is some kind metonymy going on here where the forces attributable to selection pressures becomes attributed to selection itself.
I’ve long been a bit suspicious of Dawkins’s metaphors. I tend to find Gould and Lewontin more to my liking. And while I can’t recall whether or not Dawkins has ever engaged in quite the kind of ad hominem foolery against Gould that Dennett (among others) has, I don’t know that he’s ever distanced himself from it either. The weird thing is that I’m not even sure there was truly a substantive scientific issue involved in the big Dawkins vs Gould debate. I haven’t traced it into all its details, but I’ve had the impression at times that it was just a case of “lumpers vs splitters”. That is, Dawkins — the lumper here — sought to designate all the processes involved in evolution Darwinian, and Gould wanted to keep the term Darwinian for just a certain set of core processes, but that in the end the underlying phenomena were not much in question.
Update:
From Richard Lewontin’s The Triple Helix: Gene, Organism, and Environment (pp 3-4):
It is not possible to do the work of science without using a language that is filled with metaphors. Virtually the entire body of modern science is an attempt to explain phenomena that cannot be experienced directly by human beings, by reference to forces and processes that we cannot perceive directly because they are too small, like molecules, or too vast, like the entire known universe, or the result of forces that our senses cannot detect, like electromagnetism, or the outcome of extremely complex interactions, like the coming into being of an individual organism from its conception as a fertilized egg. Such explanations, if they are to be not merely formal propositions, framed in an invented technical language, but are to appeal to the understanding of the world we have gained through ordinary experience, must necessarily involve the use of metaphorical language. [...] Indeed, the entire body of modern science rests on Descartes’s metaphor of the world as a machine which he introduced … as a way of thinking about organisms but then generalized as a way of thinking about the entire universe. [...]
While we cannot dispense with metaphors in thinking about nature, there is a great risk of confusing the metaphor with the thing of real interest. We cease to see the world as if it were like a machine and take it to be a machine. The result is that the properties we ascribe to our object of interest and the questions we ask about it reinforce the original metaphorical image and we miss aspects of the system that do not fit the metaphorical approximation. As Alexander Rosenblueth and Norbert Weiner have written, “The price of metaphor is eternal vigilance.”
me thinks thou art being hyper-picky, but i git ur drift totally. dawkins is super awesome, and can has cheezburger, but is also a pill to some at times, i believe. but yeah- you’re kinda mincing words and symantics. you gots to dumb it down some to keep it real. that said, a passive process can be the catalyst for change, so perhaps in a sense it is “driving”?
I’m with you on this one Plover. Dawkins is using a lazy metaphor and the result is language that can be misconstrued. It is similar metaphors that generated misconceptions such as social darwinism and evolution=perfection. Gould was usually more careful in how he wrote about it while still making it accessible.
Sean Carrol is the person now writing the best lay person books on evo-devo.
driving force of evolution = successful path of least resistance
linky no worky, p-bird. please be to fixit
kthxbai!
What is funny is I read this post on my RSS but I didn’t know WE had posted it. I think I see your point, but it could be construed as ploverish.
Linky been fixy
I think this metaphor – and others in this vein (I’ve heard biologists more than once talk about feature X of organism Y being “designed” to do something-or-other, which is always enough to make Mike Behe and his ilk achieve spontaneous orgasm) is an artifact of language more than anything else. Our language lends itself to assigning purpose to the purposeless. Saying “natural selection drives evolution” is sort of like talking about the moon rising and setting. Although I think our tendency to give agency to things that lack it is probably also a psychological quirk of humans – part of what makes us intelligent in ways that other animals aren’t. Assuming that other things are purposeful allows us to try to out-strategize them, because we can try to think like they would think. I think it’s an assumption that we tend to over-rely on, just because when it does work, it works so darn well that it makes up for how doofusey it makes us look when it doesn’t work.
I always kinda thought the Dawkins/Gould fight was a gene selection/group selection slapfight….but I’m a big dummy, mostly, so I’m probably wrong.
that linky made my head swim. and not just because of the cough syrup either.
that’s some ridiculously fine-toothed combing. but whatever, i know nothing about this evolution stuff
Richard Dawson is a tool. We have to deal with him (or a fan of his) every day on B4B. He goes by the name of Tyler “Fight Club” Durden.
Personally, I think he’s hot for Brad Pitt.
More like William Pitt. Whiggery!
Jillian:
Concerning the human tendency toward excessive assignment of agency, I agree completely. However, I don’t think that’s the problem here per se.
I don’t think “Saying ‘natural selection drives evolution’ is sort of like talking about the moon rising and setting.” describes what’s going on. While the idea of the moon rising and setting is not a useful framework for celestial dynamics, it is a perfectly adequate description of the moon’s activities for many purposes.
It seems to me that saying “natural selection drives evolution” is more like saying “DNA constructs proteins”: sure, DNA is an indispensible element of most protein synthesis in living systems, but its role in the process is not to construct the protein, and to say that it does obscures how the process works.
Of course, human language is imprecise, and in many situations these kinds of phrasings are “good enough”, but that strikes me as being the case largely in situations where both parties can be assumed to understand the underlying idea of what is going on. It’s also more dangerous for an abstraction like “natural selection” than it is for a more concrete item like DNA.
While I don’t think there’s much purpose in lambasting someone for using a sloppy metaphor in a random forum post, it does seem meaningful to question some of these metaphors. People have a hard enough time understanding evolution — it seems worth examining whether the language in which it is discussed contributes to the confusion.
Yeah, I agree – popular discussion of biology swims in messy, unintentional metaphor. I just wonder if it’s possible to even get rid of it if the tendency to find agency (and, incidentally, telos) in everything is as basic as it seems to be.
It’d be nice if we could get rid of it, but I don’t think it would work. Then again, I was always skeptical about the viability of Vulcans, even given Pon’far. Doesn’t mean we can’t be better at what we do, though, and it’s definitely worth examining. If we can at least develop the ability to recognize when Inappropriate Agency is happening, that would be a step forward.
It is a metagainst.
driving force of evolution = successful path of least resistance
A robust construct, it also explains middle management.
(An epilogue as an introduction: I’ve just re-read what I typed up here before submitting it. Just thinking out-loud on the internets. Not sure this post is helpful, or relevant, or meaningful in any way.)
I wonder if biology, or at least some sub-discipline within biology, could ever get to the point where metaphor is no longer used, even when explaining the science to someone unfamiliar with the field?
Metaphor has been thrown out in some pursuits, with out apology. I think mathematics is the prime example, closely followed by modern physics.
Here is a common one from physics – in classical physics the ‘fundamental source’ of electromagnetic radiation is charge acceleration. I recognize charge as the thing that zaps me from time to time. I also understand acceleration, a change in velocity over time.
In modern quantum physics the origin of electromagnetic radiation are emission events, when an electron changes from a high energy state to a lower energy state (although there are also such changes that do not emit radiation). There is an equation that defines the probability of an emission event, in a defined field over a defined amount of time. But that is it, no metaphor, nothing intuitive there – just a probability equation.
One might ask does the charge accelerate between the high energy state and low energy state, and that is why electromagnetic radiation is emitted? If we ask this question we are told not to worry about it. We are referred to the probability equation mentioned above, and we are also told even if the electron is accelerating during the emission event one could never measure the acceleration because of Heisenberg’s Uncertainty Principles. So, there is no need to worry about something we cannot measure.
If you want to know what ‘drives’ emission events, you are screwed. They just happen, and here is the probability equation – end of story, nothing more to think about.
Could biology ever be like this? Maybe someday in the future – what would that be like, I wonder?
Jump to the future–> What is the driving force of evolution? If you want to know what ‘drives’ evolution, you are screwed. It just happens, and here is the probability equation – end of story, nothing more to think about.
Back to the present–> The probability equation could be as simple or elaborate as you want, depending on how many probability distributions, with their relateted uncertainties, for various elemental parts in biological systems we come to understand in the future.
For many (if not most) of these elemental parts we will only be able to make very ‘blurry’ measurements of the distributions, which will greatly increase our uncertainty in the overall probability equation.
This could be o.k., or a disaster.
It might be o.k. – for example, there could be Heisenberg-like principles within biological systems. Maybe there is a limit to how accurately one can measure an individual cell’s mutation rate and that same cell’s generation time at the same time, because a mutation has a non-zero chance of changing the generation time, etc.
The possible disaster is that it may in fact be that our ability to make ‘blurry’ measurements will never be good enough for the probability equation to have any meaning. I think String Theory is in that pickle at the moment. We might one day have the equations in hand that completely define evolution mathematically – likely in the form of a complex mixture of probability distributions (all with there own uncertainties), but we might not ever be able to accumulate an accurate enough set of measurements of all of these distributions to test whether or not the equation is correct.
Correction:
“We might one day have the equations in hand that completely define evolution mathematically”
=
“We might one day have the equations in hand that WE THINK completely defines evolution mathematically”
A robust construct, it also explains middle management.
I hadn’t thought of it that way but it surely does.