In my initial list of Falsehoods, I included these two closely related items:
I'll make the distinction between them clear eventually. For now, let's focus on the first: Has evolution stopped for humans or not?
The reason we bring up these falsehoods to begin with is because examining them exposes assumptions or requires the investigation of knowledge that is worth talking about. In this case, the deeper question is this: What is evolution exactly? So, this is a true litmus test kind of question. With a clear understanding evolution it should be sufficient to get to the point where this question can not be answered! It can't be answered because it really shouldn't even be asked. It's like asking "OK, so the earth is a globe. Fine. But how did it get from being flat to being a globe?" That question has no answer.
Neither does "Has evolution stopped for humans?" The only species for which evolution has stopped are the extinct ones.
First a definition of evolution. There are many such definitions, and much ink has been split, pixels flashed on the screen, and breadth spewed into the atmosphere on this. It is not clear why. Biological evolution is change in allele frequency over time. This is the correct definition. Let me underscore this by giving you an alternative definition that is wrong, one that I saw on the Internet. In this particular place on the Internet, someone argued that while "change in allele frequency over time" was a nice definition, it failed to truly grasp the real meaning of evolution because this did not include, or address, "Descent with modification." That particular inter-tubule concluded that "descent with modification" is a better definition than "change in allele frequency over time." But really, that is wrong.
I assume that this was meant to refer to phenotypic change that we can normally detect or that matters with respect to the organism's interaction with the environment. Larger teeth, thinner fur, change in body size, better bodily fluids, and so on. So, with "descent with modification" if there is no physical change over time, there is no evolution over that period of time.
But in fact, most evolution is all about things not changing over time. You saw that "not" in the previous sentence, right? (Good, just checking.)
In every generation, mutations are introduced that have no phenotype correlate. You know that an allele is a version of a gene. All genes have one allele or more. If there are two different DNA sequences that are in the same locus and code for the same product (a protein) then there are two alleles of that gene. If there are three different such DNA sequences then there are three alleles, etc. In practice, there are more real alleles than there are known alleles, and this is important. (Some people will disagree with me on this, but they would be wrong IMO.) If one DNA sequence codes for two proteins that have different functions, then it is clear that there are two alleles because we can see the phenotype difference. One fruit fly has smooth eyes and one fruit fly has wrinkled eyes. (I'm mixing up fruit flies and peas on purpose ... more about that another time. Suffice it to say that I won't be using any real examples in this essay.) Anyway, everybody agrees that this is a gene with two alleles.
But what if a gene has two versions, with different DNA sequences, but both DNA sequences code for the same exact protein? Some people would say that there is only one allele, and for all practical purposes, this is OK and will not mess you up, depending on what you are doing with this information. But say you are using allelic variation to measure mutation rates. Then you know you have to make an adjustment for "silent" mutations such as this one. Or, you could simply sequence the DNA and recognize that difference means not the same, so "different DNA sequences" means that, well, the DNA sequences are different even if there are fewer proteins being coded for than there are differences.
A quick review of something you probably know, but just in case: A mutation is a change in DNA whereby one of the base pairs is changed at a particular point in the DNA sequence. There are other kinds of mutations which we shall ignore at this time (like insertions and deletions and stuff). A silent mutation is a mutation that does not have a phenotypic effect. For now we'll call a visible trait, such as blue eye color or tendency to be a Republican, a trait, and not worry about the small technicality that if you were really really tiny and could look at a protein and see altered base pair that is not expressed at the bigger scale (like eye color or voting habits).
A mutation can be silent because it changes a base pair in a codon that does not in turn change the amino acid that codon specifies. Proteins are made up of a sequence of amino acids, and "coding for a protein" means having a series of codons (three base pairs, a kind of genetic "word") that specifies the series of amino acids that the little fairies that live inside your cells will put together to make the proteins. (I'm leaving out some details and substituting said details with the word "little fairies" ... I figure it will help my search engine score on Google.)
You probably already know that many different codons (there are 64 in all) code for a smaller number of amino acids (about 20). Like skillet and frying pan both mean the same thing, so do many sets of codons mean the same thing ... the same amino acid.
A mutation can also be silent because it does change which amino acid is being placed in that position while making the protein, but in a way that does not matter. For example, if the phenotype relies on a certain part of the protein working a certain way, then the other part .. where the hypothetical mutation happened ... can vary within a certain range and the protein still acts the same.
Like screwdrivers. If you change the bit end of a screwdrivers you may have screwed yourself out of a screwdriver that works. But if you slightly change the color or size of the handle, who cares? So a mutation can be silent because it is a "who cares" mutation. This can be taken to several levels. You can get a protein that has a real functional difference, but it is one protein of seven that essentially serve the same purpose because of some kind of crazy redundancy thing and no one notices. For instance, maybe this is a protein that is essential in the synthesis of a really important molecule, but in an organism that consumes copious quantities of this same molecule found in nature in usable form. So the inability to synthesize the molecule will not be a visible trait until it is noticed by someone, like when certain crazy primates started to sail around in the ocean for months at a time and started getting scurvy because they did not synthesize Vitamin C because an ancestral primate got this then-silent mutation. The mutation never got noticed because all the other monkeys and apes were sitting around eating copious quantities of Vitamin C (fruit, vegetables, fresh blood, etc.). But these boat-monkeys were eating dried pork rind or something else without Vitamin C, so they got quite ill. No one noticed for tens of millions of years in that case! So a lot of what some people might call "neutral evolution" is going on all the time in the background with nobody caring and nobody noticing.
That is change in allele frequency over time. It is not decent with modification in the usual sense of the term.
Now, suppose something gets noticed? If an allele arises that causes a change in phenotype that matters, then it is possible that selection will act on it. Actually, I've misspoken. When I said "that matters" I meant an allele that confers a difference subject to selection. It will be selected against or for depending on the interface between the functional change and the environment. Usually against. (All change is bad, right?) But selection comes and goes and not everything is given the same ranking in the selective milieu. In other words, a single allele change with nothing else happening in a population or even a lineage is very unlikely. Usually in a population at any given time there are a bunch of novel alleles. Let's say that there is a one in a thousand chance that a single mutation will have a meaningful effect. (This is quite possibly a gross underestimate). Depending on who you believe, this could mean that in a population of 100,000 individuals, there are about 200 different meaningful alleles introduced every generation. Many populations of larger animals are probably much smaller than this. Many populations of plants probably inter pollinate among millions of individuals. So this thought experiment is pretty limiting. But just figure that it is quite possible that when you think of a "species" there may be many novelties introduced in any one generation.
Most of these meaningful changes probably have weak effects. Being a winner or loser in the mating game, for instance, may not mean you are the mother of all future individuals or an evolutionary dead end. It probably just means that you will have a little more or a little less of a contribution to future generations. Then the genes get all mixed up each generation because of a crazy little thing called love, and so on. So novel mutations that are under selection may be around for many generations. If hundreds or thousands of novelties are introduced every generation and most get to stay around for several generations, that means that there are anywhere from scads to zillions (to be exact) of mutations in a given "species" (population) at any one point of time. So now you get the picture of a constant ongoing interconnected complex mess of a situation. Dawkins' river out of Eden.
And all this time, it may be that many different selective forces are working on the phenotype. How broad and long is the tail of an alligator or a crocodile? What genes cause that particular feature? Over time, it turns out that in most systems, the genes that underlie particular major features in related forms (like all crodillians) stay the same. (But not always ... the subject of another discussion, another time.) But it may be that the particular combination of alleles that produce that phenotype vary. Here's why: There is no gene that affects only one trait, and there is no trait that is affected by only one gene (and I'm only slightly exaggerating). Therefore, the size and shape of the tail of a crocodile may be affected by, say, a dozen genes, each with a range of alleles, that in combination come up with a particular tail size and shape, but whereby that same size and shape could be obtained with several different combinations.
A metaphor is in order. Imagine starting, say, an Internet hosting company, with servers and such. You hire 12 people to run the company. They provide sales, technical maintenance, support, make purchasing decisions for new equipment, brew coffee and change light bulbs, etc. They have something like 50 or 60 jobs divided among the 12 of them to get the whole bigger job of running this company done.
Now clone the company so that there are 100 identical companies, and let them run without communicating with each other for five years. At the end of ten years assume you still have the same 100 companies that are doing the same thing. Now interview each person and see what their job is exactly. Are the 60 specific jobs divided among them being done by exactly the same nominal positions in the company as when you first started it up and cloned them? Many key jobs probably are. You need a Unix Geek to do the Unix stuff, and your sales rep is probably not that person. But many jobs will have migrated back and forth among different categories of individuals. Who makes the coffee, who has input into web design, etc. etc. could all shift around quite a bit. You may end up with some very different models for internal functioning among these 100 companies that are all doing the same thing on the outside. In other words, you may end up with 100 different combinations of alleles all making the same exact phenotype.
This is a LOT of change in allele frequency over time. It is ZERO descent with modification.
OK, that is how evolution works. Now let's get back to the original question: "Has evolution stopped for humans." I feel like I don't even want to answer that question at this point, don't you agree? Are humans organisms that use genes to encode protein sequences and pass them on over time? Do humans have phenotypes coded for by complex interactions of many many genes? And so on?
How can evolution stop? What would that mean? It means nothing other than extinction and at the moment we are not extinct. So we are evolving. Evolution has not stopped for humans.
So, if we measure phenotypic change in humans over time and see some change, then that must mean we are evolving. But, if we measure phenotypic change in humans over time and don't see any, then that does not mean we are not evolving. We are still evolving even if there is no phenotypic change.
There are many other aspects to this question, which I'll explore later. For now, I'll just briefly touch on the second iteration mentioned above: The fallacy that serious scientists ever consider that evolution has stopped/not stopped for humans. By now you probably understand what this is about.
Now and then you will see a headline in the science section of the NY Times or somewhere that says "Scientists show that evolution has not stopped for humans after all!" or words to that effect. There are two things going on here. One is the press playing off the fallacy that evolution has stopped for humans. The other is that some scientists who is probably not really a biologist, or certainly not a population geneticist, woke up one day and realized that evolution has not stopped for humans. No serious scientists who actually studies human evolution would ever have this thought.
Above I talk only about neutral evolution. This does not mean that I don't think meaningful, adaptive change is not happening now with humans. I'm just not talking about that in this post. Another time.
There is also an interesting discussion to be had as to why humans tend to gravitate towards presuming the evolution has stopped for our species. The reasons are often not just born of ignorance, but can be rather nefarious. As you might expect given that we evolved from apes ...
More Falsehoods !!!
This post is one of a series on the topic of falsehoods. The following is a list of falsehoods posts in order:
- The Falsehoods
- "False Pearls before Real Swine"
- Falsehood: A baby is not the biological offspring of its adoptive mother
- Falsehoods: Has evolution stopped for humans?
- Natural Selection is Survival Of the Fittest (A Falsehood)
- Falsehood: Nature maintains balance.
- Is it a Falsehood that Humans Evolve from Apes?
- The poor and the dark skinned have more babies than the rich and the light skinned
- Acting for the survival of the species (a falsehood)
- Culture Overrides Biology (Another falsehood)
- What is the Placebo Effect, and it it getting stronger?
That explanation, of course, points out how we understand things differently than did Darwin and other biologists in the decades following, who didn't know what an allele was.
What's the allele for endosymbiosis?
I think it was Lynn Margulis who raised this objection against equating evolution with allele change, and I also think it's pretty much a show-stopper.
toto: I can't think of a more profound example of change in allele frequency than the combining of two previously distinct genomes.
There is no show stopper here. There is merely the adherence to the trivial and sophistic. Let the show begin.
I think I've only ever heard this argument made from people who mistake our dampening of natural selection for a wholesale stopping of evolution. The only argument that can be made is that we are, through health care and social support structures, keeping alive the more unfortunate parts of society so their genes can theoretically be passed on. This obviously has nothing to do with "stopping" evolution -- just in mitigating some of the selection factors.
the only argument that can be made is that we are, through health care and social support structures, keeping alive the more unfortunate parts of society so their genes can theoretically be passed on.
A falsehood we will also deal with in time!
Nice post, Greg, I really don't have much to add in the way of substance, but a neat little bit about human evolution that you already probably know: lactase persistence allelic frequency in humans makes an excellent example of how humans are, in fact, still evolving. They also demonstrate the historical shift in allelic frequency as well as a strong selective event in Northern Europe for lactase persistence. These allelic frequencies are still changing and will continue to change for quite some time...
A falsehood we will also deal with in time!
Awww. I suppose you want me to eat my peas first.
Let's restate the question about human evolution, and pick a trait that used to have selective advantage but to the extent that it is correctable in industrialized society it isn't very selective anymore: good eyesight.
Should we expect more and more combinations of alleles that allow for poor eyesight to become common or even prevalent over the much smaller number of allowed combinations that allow for good eyesight? Between mutation and drift, how many generations would it take for a noticeable shift to become apparent? Tens? Hundreds?
So I'm still wondering, are we different from the Cro-Magnon in any meaningful ways?
But why is it ONLY a change in allele frequency over time?
Don't changes outside of genes count? Epigenetic changes or changes in DNA switches should count also, shouldn't they?
I can see a case being made for ANY changes in DNA over time being the definition of evolution, including changes in 'superfluous' DNA.
New England Bob: You are absolutely correct. The "change in allele frequency" idea only really works in a world in which genes code for proteins and everything is a protein, and this totally ignores non-genetic transmission.
One person's answer to a fallacy is another person's fallacy.
The contrast in this post is really about phenotypic vs. genetic change. The points you bring up are all to be addressed as well.
Wait...does this tie into the biological v. genetic parent post too? Is this a series? Link them up if they are, please!
It is a series. I'm working on the linking up now..
Okay, and thanks! *sigh* It took me forever to find the bit of the last aricle I was wondering about when paragraph the whichever twigged my brain and told me "maybe you'd better go look back at X..."
Now let's get back to the original question: "Has evolution stopped for humans." I feel like I don't even want to answer that question at this point, don't you agree?
The way I see it, if someone asks you a painfully ignorant question, you can either suck it up and answer it yourself, or let some dumbass with an agenda and no grasp of the topic whatsoever answer it for you.
Brother: I think you may have missed the point by about ten miles...
Brother: I think you may have missed the point by about ten miles...
To be fair, I was letting a stray thought out leak of my head, not commenting on the article.
I can imagine a hypothetical scenario where evolution might stop for a species that is not extinct.
One needs to posit zero variation (every allele is at 100% frequency - it's a population of perfect clones), and a mutation rate of zero (a mechanism for replicating the genetic information that is 100% perfect).
Hard to say how long such a species would actually last in the real and ever changing world, or whether something like a zero mutation rate is even physically possible.
I can't think of a more profound example of change in allele frequency than the combining of two previously distinct genomes.
Then you have a non-standard definition for 'allele'. NTTIAWWT.
Evolution is a change in frequency of heritable features, in general ("feature" here really means anything physical, not just the phenotype - a certain nucleotide at a certain locus is a feature).
Heredity needs not be mediated by genes, in principle. Evolution and natural selection don't care how heredity is mediated, as long as it's actually there. I mentioned endosymbiosis, others have metioned epigenetics. Another striking example is in origins-of-life models, where primitive chemical replicators can maintain some heredity (and thus evolve) without any explicit repository of genetic information. This also occurs in artificial evolutionary systems, which may or may not use a genome, but nevertheless fully utilise the power of evolution by natural selection.
Of course, in practice, in modern lifeforms, almost (almost) all of heredity is mediated by genes, and therefore gene-based evolution represents 99.999% of situations of interest. But it's nice to be able to separate a general principle, from a particular implementation.
I agree with you guys..
It strikes me that change in gene frequency over time is a fairly slow adaptive process. And that we live in a world of very rapid environmental change. Changes in gene-mediated fitness are simply too slow to keep pace. Cultural evolution, rather than changes in gene nature or distribution, is the dominant force in human affairs.
>> the only argument that can be made is that we are, through health care and social support structures, keeping alive the more unfortunate parts of society so their genes can theoretically be passed on.
> A falsehood we will also deal with in time!
You don't think the burden of deleterious mutations should probably rise when the variance in the number of surviving children per individual goes down? I won't say it's a fact, but it seems highly likely. I'd be pretty impressed if you could demonstrate that it's definitely false or even just unlikely.
Eric, there are a number of reasons that this is a falsehood, and I'm afraid that one of them is that our system of medical care is deteriorating while the complexity and severity of our medical problems is increasing, so .... the whole argument turns around. But that is not the main reason. Stay tuned.
But what if a gene has two versions, with different DNA sequences, but both DNA sequences code for the same exact protein? Some people would say that there is only one allele, and for all practical purposes, this is OK and will not mess you up, depending on what you are doing with this information.
Couldn't such a situation still have a phenotypic effect? For example, we know that the way genes are expressed in cells is often determined by small molecules such as microRNA's. So theoretically, two alleles with different DNA (and hence corresponding RNA) sequences but coding for the same protein, could be expressed differently in cells if the section where they differ is a binding site for a micoRNA. I don't know if there are any known cases of this, but it seems perfectly plausible to me.
Jon: are we different from the Cro-Magnon in any meaningful ways?
For one, no Cro-Magnon ever could have asked such a question. I am not being facetious: our culture is as much a part of our evolutionary heritage as the DNA, although it is not transmitted in the form of alleles. (It is, at the same time, an essential part of the environment within which the genetic bits evolve, but so is every gene a part of the environment within which each other gene evolves, so that's no distinction.)
This makes trouble for the definition of evolution, unless you enlarge the definition of "alleles" to include cultural phenomena, with some danger of turning the latter into mush. But evolution doesn't care about our definitions; it's for us to try to keep up as best we can. Anyway trying to understand human evolution in isolation from culture is folly.
I wonder if there is something wrong with the way evolution is taught in schools (it's still taught in the USA, isn't it?) When evolution is mentioned people seem to think about the phenotype changes only but don't seem to understand that the process of differentiation of species occurs over unbelievably long periods. We can cross-breed some species with mixed results on the viability and reproductive capacity of the offspring - for example we cross a donkey and mare to get a mule (does anyone still call it an 'ass'?) - anyway, the phenotype changes are obvious and yet the horse and donkey have not differentiated enough that no offspring will be produced.
The other thing about the public perception of evolution, and this is something that really annoys me, is that people imagine that some of our ancestor species are still extant. People have these bizarre notions that the apes developed from the modern monkeys and that humans developed from one of the (other) modern great apes. (And incidentally, people don't seem to realize that humans are one of the great apes.)
So in general, the public's understanding of evolution seems to be screwed up beyond belief. It's no wonder that they easily fall for claims that humans are no longer evolving. Of course for some creationists who pretend to believe in evolution that is the case - modern humans are god's planned species. More sophisticated creationists like Francis Collins might say that human evolution was inevitable but that god may have a plan for further human evolution.
The only place I've seen this argument lately is as a strawman. A swarm of racists (or sockpuppets of one racist) in this Pharyngula thread claim that "leftists" believe "evolution by natural selection somehow stopped for humans 50,000-100,000 years ago" but since this is clearly false, black people are inferior.
moar: I think the "evolution stopped" argument is pretty popular among eugenicists. They feel that with natural selection failing to hold up its end, it's up to them to provide artificial selection to fill the gap.
I confess that as a teenager I thought my myopia was pretty strong support for the first part of the above argument. I reasoned, though, that in no time (geologically speaking) we would have full control of our genomes and could then replace unfortunate alleles directly in our germ plasm, as desired. Eugenicists don't make this connection because it doesn't fit with what they really want to do, which is to practice eugenics, i.e. treat people like farm animals; any external purpose is more or less incidental.
* * *
The "descent with modification" expression is properly not considered a definition, but rather just identifies the circumstances where evolution necessarily occurs. A definition of evolution that depend on alleles is limiting because evolutionary natural selection can occur in circumstances where there is nothing that could be pointed to acting in the role of an allele.
As I see it, "descent with modification" is the unavoidable consequence of evolution, rather than the definition of it.
Eric, if selection pressure is reduced in way such that a greater variation of offspring survive in every generation, the so-called burden of deleterious mutations will not change, because by definition the extra surviving variants would not be deleterious anymore, because the environment has changed and the selection pressure that once made them deleterious is no longer present. (Not to say that these same variants might not once again become deleterious - or even advantageous, when the environment inevitably changes yet again.
moar: This comes directly from the previous paper and book by Henry Harpending and others demonstrating not only post-pleistocene evolution, but suggesting (rightly or wrongly) accelerated evolution in humans. There are some ways to interpret that as suggesting that humans have divided into distinct populations that are becoming increasingly different and that eventually we'll have several species: A low-intelligence dark skinned laboring species that is a little dangeriou, a slanty-eyed yellow highly intelligent species and a white species with exceedingly high moral standards which will enslave the dark species and try very hard to kill off the yellow species before they kill off the whites.
Harpending et al are correct in that there is plenty of evolution. I think they have quaint views of both ancient and modern population dynamics, and that's OK because I will correct that by and by, because it takes time. There is a lot of evolutionary change going on in humans (it is impossible to imagine how that would not be the case) although I have not addressed that in this post. I am just now putting the finishing touches on a post that will begin to address it, which will be posted a bit later on this week.
For clarity's sake, can the triple negative in the quote below be simplified to say the following?
This does not mean that I don't think meaningful, adaptive change is not happening now with humans.
Regarding the actual meat of the article, I think a good case has been made for the fact that asking whether we've stopped passing on mutations is like asking whether the Earth has stopped spinning.
But, when people ask if evolution has stopped for humans, I don't think they're really asking if we've stopped passing on mutations. In colloquial verbiage, the question is really asking whether we've removed certain selective pressures, to our detriment or to stunting of our development.
In other words, the concern of the question is in whether we as a population are more vulnerable to phenotypic degradation now that nature is less able to burden the phenotypically "unfit" to the point of reduced reproductive success, and whether beneficial alterations have a fighting chance to proliferate.
Needless to say that the question has no simple answers, so I look forward to the next instalment.
I recently read an article explaining that synonymous mutations can sometimes affect the phenotype by altering the splicing of mRNA. I can't remember where it was, unfortunately.
I've heard people claim that, due to increased health care, etc... natural selection doesn't apply to humans anymore. My reaction always is that he (or she) comes from that one sperm cell which beat several million others to the egg. That's quite some selection.
I enjoyed the post and agree with it, but I would like to point out (what I believe is) a mistake.
"If we measure phenotypic change in humans over time and see some change, then that must mean we are evolving." Not necessarily. Phenotypic changes can be caused by environmental or cultural changes. An example would be the increased height of Americans over the past hundred years.