A recent research study concludes that during the nineteenth century polygamy hurt the evolutionary fitness of Mormon wives:
Polygamy practiced by some 19th century Mormon men had the curious effect of suppressing the overall offspring numbers of Mormon women in plural marriages, say scientists from Indiana University Bloomington and three other institutions in the March 2011 issue of Evolution and Human Behavior.
Simply put, the more sister-wives a Mormon woman had, the fewer children she was likely to produce.
“Although it’s great in terms of number of children for successful males to have harems, the data show that for every new woman added to a male’s household, the number each wife produced goes down by one child or so,” said IU Bloomington evolutionary biologist Michael Wade, whose theoretical work guided the study. “This regression is known as a ‘Bateman gradient,’ named after the geneticist who first observed a similar phenomenon in fruit flies.”
There are so many unsupported (and possibly unsupportable) assumptions in this paper that its hard to know where to begin. For instance, the researchers assume that evolutionary fitness is directly correlated to the number of children. But the relative gains that can be had by increasing the number of children can be offset by the effects of nurture on ensuring a particular child’s survival. Having more mothers around can certainly help to ensure a child doesn’t get killed before it’s their time to have children of their own. In other words, what matters is not how many total children that an individual spawns but rather how many survide to reproduce the species.
But even their assumptions about how many children would be produced are questionable. As the late Australian philosopher David Stove once noted,
Do you know of even one human being who ever had as many descendants as he or she could have had? And yet Darwinism says that every single one of us does. For there can clearly be no question of Darwinism making an exception of man, without openly contradicting itself. “Every single organic being”, or “each organic being”: this means you.
Since no woman bears as many children as is naturally possible, any claims about having more or fewer children must be based on some relative standard. So what was the basis for comparison? Mormon women who were not in a polygamous marriage.
Although I think the research is so flawed that the results are all but meaningless, let’s assume that it is true. What can we extrapolate from the conclusion? The obvious parallel would be women who take contraceptives compared to those who do not. Since the former are statistically less likely, all other things being equal, to have children than the latter, we can say that contraceptives hurt the evolutionary fitness of modern women.
Taking birth control therefore ensures that some other women’s children are likely to carry on the human race.
(Via: Stephen Windham)
May 6th, 2011 | 11:17 am
Why was the outcome of this study so surprising? Adults eat and drink more than children, and require more space, so adding an adult to a household costs more than adding a child. Every instant spent by a mother or sister-mother (is that the right term?) caring for a child is a moment not spent in generating income (whether money, or in some subsistence way). Ergo, adding a largely non-productive (in this sense; as a parent of three, soon to be four children, I of course realize that caregiving is “productive” in a more general sense) adult to a household means less resources are available to add children.
May 6th, 2011 | 11:19 am
“Do you know of even one human being who ever had as many descendants as he or she could have had? And yet Darwinism says that every single one of us does.”
Well, yes, of course, tautologies are always perfectly true, and the methodology here makes it a tautology — however many children you had, that’s how many you were “fit” to have when all the physical AND social AND whatever-else variables are accounted for. That’s why this kind of speculation about “fitness” is fairly useless, at least on this level.
May 6th, 2011 | 11:36 am
Well, the actual paper (http://stoppolygamyincanada.files.wordpress.com/2011/02/moorad-et-al-2010-evol-hum-behav-3.pdf) makes clear that it’s only talking about and measuring reproductive success in terms of number of offspring. It is not claiming that’s the only possible measure, that ‘quality of offspring’ has no effect. It’s simply measuring what the available data allows measuring.
Why, exactly, would that not be a valid basis for comparison? The data did include polygamous and single marriages.
Maybe “Darwinism” does. Actual evolutionary theory doesn’t, and takes things like contributions of rearing offspring into account. That just wasn’t the focus of this paper drawing from this data set, that’s all.
May 6th, 2011 | 11:48 am
Ray Ingles It is not claiming that’s the only possible measure, that ‘quality of offspring’ has no effect. It’s simply measuring what the available data allows measuring.
That’s like saying, “The moon is the brightest object in the sky. Well, yes, except for the sun.
When you leave out a key factor that has a significant impact on the results, then its fair to say that the results are rather limited and not very interesting.
Why, exactly, would that not be a valid basis for comparison?
Because it’s too subjective to be meaningful. Depending on your basis of comparison, the result could be exactly the opposite. For example, if you compare these polygamous wives to Mormon women who never marry or have children then it turns out that polygamy helps evolutionary fitness.
Actual evolutionary theory doesn’t, and takes things like contributions of rearing offspring into account. That just wasn’t the focus of this paper drawing from this data set, that’s all.
So actual evolutionary theory takes such factor into account except for . . . when they don’t? That makes no sense. Whether it was the focus of the paper is irrelevant. The fact that such a variable can radically change and even nullify the results is not something that can be waved away.
May 6th, 2011 | 11:57 am
[...] How Contraceptives are Like Polygamy [...]
May 6th, 2011 | 12:04 pm
Ray Ingles:
that’s a great line
“It’s simply measuring what the available data allows measuring.”
That encapsulates the fundamental failure of so called “human sciences:” it has to reduce reality to what can be measured. Except, our experience of reality is much broader than measurability.
But the real disaster, is that most “human scientists” I know end up quietly taking for granted that what cannot be measured simply does not exist, and that their “science” describes “reality.”
May 6th, 2011 | 12:28 pm
Joe Carter –
A paper can’t address all possible questions! The authors were trying to isolate a specific effect of a specific practice. In other words, all other things being equal, a Mormon woman in a single marriage would have more children than a Mormon woman in a polygamous marriage. And a Mormon male in a polygamous marriage had more children compared to a man in a monogamous marriage, both due to number of mates and extended ‘reproductive tenure’, the number of years he sired children.
The demographic data didn’t allow comparing, say, how well the children of polygamous marriages reproduced relative to the children of monogamous marriages – which might indirectly measure aspects of ‘quality of offspring’. If you want to check that out, I suppose some detailed digging into a lot of genealogies might be a fruitful pursuit for you…
Why do you think this one paper is all of evolutionary theory?
Scientific papers of necessity have a narrow focus on a particular issue. Imagine if I disputed a paper measuring the resistance of a particular semiconductor material by saying, “But real-life circuits have to worry about induction and capacitance, as well as heat dissipation!”
May 6th, 2011 | 12:29 pm
Carlo Lancellotti –
So… we shouldn’t measure what can be measured?
May 6th, 2011 | 12:41 pm
The obvious parallel would be women who take contraceptives compared to those who do not.
No, the obvious parallel is women who limit the number of children they have, by any means. Those who advocate Natural Family Planning do not consider it contraception, but they do consider it as effective in limiting and spacing children as the pill.
Taking birth control therefore ensures that some other women’s children are likely to carry on the human race.
Mandatory celibacy for Catholic priests, and vows of celibacy or chastity for nuns, monks, brothers, and so on, make sure that they have no children to carry on the human race. Are we to be concerned that the resultant evolutionary effect will be to diminish whatever heritable characteristics dispose people to serious religious commitment?
I think the whole topic is nonsense, and it sounds a bit like eugenics.
May 6th, 2011 | 12:41 pm
“So… we shouldn’t measure what can be measured?”
We should realize that the limitation of what can be measured may be distorting the true picture of reality, since it excludes some unmeasurable, and possibly highly significant, factors from any given situation.
But sure, measure it.
May 6th, 2011 | 12:52 pm
Ray Ingles A paper can’t address all possible questions! The authors were trying to isolate a specific effect of a specific practice. In other words, all other things being equal, a Mormon woman in a single marriage would have more children than a Mormon woman in a polygamous marriage.
If this is really all the paper is saying then it should have never been published. If you ignore the most significant factors that totally invalidate your conclusion then it isn’t really scientifically noteworthy.
Why do you think this one paper is all of evolutionary theory?
Where did I say it was?
Imagine if I disputed a paper measuring the resistance of a particular semiconductor material by saying, “But real-life circuits have to worry about induction and capacitance, as well as heat dissipation!”
But real-life circuits do have to worry about those variables. If you are writing about some fantasy-world semiconductor then I guess you can say that those things can be left out. But if you are creating a semiconductor for the real world those factors have to be accounted for.
David Nickol No, the obvious parallel is women who limit the number of children they have, by any means.
Well, yes, in reality that would be the parallel. But in the goofy standards of this paper we are only allowed to make a comparison with one cohort. The group that is more likely to be able to control birthrates (statistically speaking) is the one that takes contraceptives.
Are we to be concerned that the resultant evolutionary effect will be to diminish whatever heritable characteristics dispose people to serious religious commitment?
For the record, I think all this “evolutionary fitness” stuff when applied to humans is nonsense.
May 6th, 2011 | 1:48 pm
“. . . we can say that contraceptives hurt the evolutionary fitness of modern women.”
And men. Genetic recombination is thwarted.
May 6th, 2011 | 1:51 pm
To second Ray’s point, it’s not optimal from an evolutionary POV for an animal to utilize every possible opportunity for breeding, and interestingly many animals actually do set up barriers to breeding…for example females will only breed when a male wins in some competition against other males. This is sensible from an evolutionary standpoint because if a population is under some type of environmental stress (say a famine), delaying breeding is a good trait to be able to employ.
In fact, the strategic goal is to get traits successfully passed down. One way to do this is to have ‘as many children as possible’, playing the numbers so to speak. Another strategy, though, is to have fewer kids but invest more resources into them. Some animals deploy the numbers game, fish for example that lay thousands of eggs…others deploy the investment strategy, esp. mammels.
When you’re using the investment strategy, timing is everything. Since timing matters, you want to be able to conceive when the time is right. Hence of course nobody has as many children as they are biologically capable of having. There’s a surplus of “conception capacity” built into the system because when the opportunity presents itself, you want to be good to go.
In terms of polygamy, then, this shouldn’t be very surprising. Polygamy works out fine in evolutionary terms for the ‘alpha male’, but it does mean that other males will be ‘wasted’. Not only that, it means that children are wasted as well since half the children will be males, but a polygamous society needs more females than males. So overall the rewards for children are less because half of them will take resources to raise and become problems when they come of age and will compete for wives. Also evolution works on multiple generations, not individuals. In one generation you may have a strong man who can command many wives, by generation two he’s dead. If he tries to pass on the title of ‘strong man’ to his children (which would be a good ‘evolutionary impulse’), the stability of the position is very dubious *unless* the ‘strong man’ discourages polygamy for everyone else thereby making the male.v.male competition for wives less high stakes and intense. I can see how this could evolve into a kingship system where the king gets multiple wives but everyone else expects only one. Otherwise a polygamous species is going to be unlikely unless you also had a different ratio of male-female births than 50-50.
Joe
That’s like saying, “The moon is the brightest object in the sky. Well, yes, except for the sun.
No it’s more like saying one particular star (not counting the sun) is the brightest object in the night sky in visible light but another particular star is the brightest in another part of the spectrum.
Because it’s too subjective to be meaningful. Depending on your basis of comparison, the result could be exactly the opposite. For example, if you compare these polygamous wives to Mormon women who never marry or have children then it turns out that polygamy helps evolutionary fitness
Subjective? You’re comparing the number of children non-polygamous women have to the number that polygamous Mormon women have. Most women marry and most have children and presumably Mormon women have even higher percentages of childbirth and marriage. No doubt in the sampling there are some small portion of women who marry (polygamous and non-polygamous) who never have children. They would likewise impact the averages so there’s no subjectivity problem there. Women who don’t marry may be a problem if they represent a large portion of the population in question. I would imagine, though, in traditional Mormon communities this is a very small portion of the population.
If this is really all the paper is saying then it should have never been published. If you ignore the most significant factors that totally invalidate your conclusion then it isn’t really scientifically noteworthy.
How do you know which factors are the most and least significant? If ‘quality of offspring’ is a highly significant factor here, then those born of polygamous marriages would be markably better than those born of monogamous marriages. Is there any evidence that this is true by any reasonable metric? If not then it seems to be a safe working assumption that the quality of the average individual offspring is roughly equal thereby making numbers the most significant factor.
A fair criticism of the paper is that evaluating ‘evolutionary fitness’ by looking at one, two or three generations is like trying to predict the result of a football game by looking at only a few plays and nothing else. You can look at, say, three plays and observe that yes team A appears stronger than team B, but that’s hardly enough to really trust a prediction that A would win over the entire game. Even though Mormons are excellent record keepers, we only really have what, maybe 4 good generations where polygamy was recorded in detail?
Still it’s not entirely worthless to watch 3-4 plays and ask what team does a better job advancing the ball.
David
the obvious parallel is women who limit the number of children they have, by any means. Those who advocate Natural Family Planning do not consider it contraception, but they do consider it as effective in limiting and spacing children as the pill.
This might hint as to why some women like polygamy. In a culture and/or religion where women are expected to have many children, polygamy may be a way for the woman who doesn’t want to have a lot of children to have fewer without signalling that she is going against social norms. Also from an evolutionary POV it may work with the ‘investment strategy’. If she has fewer children and her husband has many more, she has relatively much more influence on her own children than her husband does as he must divide his time and attention.
Pentamom
e should realize that the limitation of what can be measured may be distorting the true picture of reality, since it excludes some unmeasurable, and possibly highly significant, factors from any given situation.
Yes yes but Joe’s criticism is unfair….and that’s not surprising as he’s usually a bit unhinged whenever the topic’s evolution. Can you study the ‘fitness’ of offspring in humans? Yes, but there’s numerous metrics you can use (lifetime income, average educational attainment, # of grandchildren they had, incidence of mental illness or criminality etc.). You’d probably need bigger samples and much more detailed records. A worthy topic if someone wants to undertake it but that doesn’t mean the ‘easier’ subject of research should be knocked. If we wanted to apply Joe’s criticism equally, we should say that people should stop citing the Bible regarding Jesus. After all, the Bible leaves huge parts of Jesus’s life in silience. By Joe’s logic then, since we can’t learn as much about Jesus’s childhood from the Bible as his later preaching, then we should just not bother at all. But I don’t expect Joe to be consistent there….
May 6th, 2011 | 2:03 pm
Boonton In fact, the strategic goal is to get traits successfully passed down.
No it doesn’t. Evolution (under your view as an atheist) has not intention, not strategy, not direction. Please try to make your claims without resorting to teleological language.
May 6th, 2011 | 2:18 pm
First didn’t say my view was an atheist.
Second, you raise a valid point about language. But not too much. Look if you had a leak in your house, you’d try to track it by thinking wherever the water is escaping from the pipe, it will ‘want’ to find its way downhill. Therefore if you find wetness on the first floor, the second floor, the third floor but not the fourth, you’d conclude there’s a leak somewhere on the 3rd and the water is coming down. You wouldn’t conclude that the leak is in the basement and somehow traveling up to the 3rd. But, of course, the water itself doesn’t ‘want’ to go down and the earth’s gravity doesn’t ‘want’ to ruin your building by pulling the water down. But the ‘teleological language’ is useful because it cuts out a lot of this long winded explanatory text.
So species A has lots of offspring and invests littl etime in them. Species B has few but puts a lot of energy into each one. Species A & B aren’t ‘choosing strategies’ in persuit of a goal, but they are nonetheless utilizing different strategies and the fact is unless both strategies happen to be exactly equal in terms of long run success, one will be better htan the other and that will have consquences. (That’s assuming, of course, everything else stays the same…A might be at a disadvantage until a change in environment turns the tables).
May 6th, 2011 | 2:58 pm
Joe Carter –
They are able to actually measure and quantify the effect, which allows them to rather specifically define how strong a countervailing effect would have to be to ‘invalidate’ it.
And other scientists measure those properties. The resistance of a material is an important variable. Lots of lab setups are impractical simply because they are trying to tease out very specific variables.
And I’m not seeing evidence yet that you actually looked at the paper itself rather than the popularizing article about the paper. I submit this for your consideration: http://www.phdcomics.com/comics/archive.php?comicid=1174
And you said to Boonton:
Well, I would, but HTML doesn’t really do math very well. Guess we’ll have to settle for English and metaphors with caveats. Sorry.
May 6th, 2011 | 3:05 pm
The obvious parallel would be women who take contraceptives compared to those who do not. Since the former are statistically less likely, all other things being equal, to have children than the latter, we can say that contraceptives hurt the evolutionary fitness of modern women.
Do we know this, really? Leave out abstinence as a contraceptive. All things being equal, an average woman having sex with contraception will not conceive at the rate of an average woman having sex without. But what if women who use contraception have more sex? I don’t have the book in front of me but I recall the chapter on abortion and crime in Freakonomics had some research to indicate that on average lifelong fertility in women wasn’t altered by Roe. In other words, the availability of abortion didn’t actually decrease the number of children the average woman will give birth too in her lifetime. In other words, if the average woman has two children, the only difference abortion typically makes is that those kids may come at ages 19 and 23 without abortion, 25 and 29 with. There’s no much reason to think that contraception is any different. If women who use contraception in their lifetimes simply end up having more sex earlier, later or inbetween than women who don’t, then the # of children isn’t going to change.
This is an obvious parallel for more research but that’s not Joe’s thing so he just assumes there’s a difference when in reality no one here knows unless/until they actually go and find out.
May 6th, 2011 | 4:01 pm
Before the pill was widely available and affordable, the TFR (total fertility rate) for the US peaked at about 3.6 children per female (1959). However, by the mid 1970s, the TFR dropped to 1.8 before reaching a nadir of 1.5 in 1980. If abortions can be also classified as contraception, we can surmise there is some statistical correlation between ABC and fertility. Today, the TFR has recovered somewhat mainly due to immigration (both legal and illegal). Today, the TFR in the US is about 2.06 births per female. But subtract Hispanics and it drops to 1.6 births per female.
May 6th, 2011 | 4:53 pm
I’m totally at a loss as to what Mr. Carter is objecting to. He doesn’t object to the fact that there is popular discussion of the study, which is the only place his criticisms have any purchase: yes, it’s a very narrow point, the comparison is one out of many possible comparisons, etc. Science works, most of the time, by breaking things into small, manageable problems so that things can be measured. Granted, I don’t know why the general public should give two figs for the study, taken by itself, but if scientists only published results that were super-meaty and immediately interesting to everybody, science would be far less successful.
Maybe it’s the popular connotations of “evolutionary fitness” that are objectionable? But all that really means in the study, I take it, is that they are likely to have a smaller set of descendants in future generations than monogamous women of their time. That’s not meaningless – it’s hardly something to write home about, either, but it’s not meaningless or hopelessly confused.
May 6th, 2011 | 5:41 pm
Maybe it’s the popular connotations of “evolutionary fitness” that are objectionable? But all that really means in the study, I take it, is that they are likely to have a smaller set of descendants in future generations than monogamous women of their time. That’s not meaningless – it’s hardly something to write home about, either, but it’s not meaningless or hopelessly confused.
I suppose this is the ‘teleological language’ he’s fighting. The implication that there’s something bad about being ‘less fit’ in terms of evolution. But this is where philosophical baggage is being added to a scientific theory for no good reason.
To use the analogy I used before, one says “water wants to travel downwards” hence the puddle in the basement is due to a leak on some floor above the basement. That doesn’t make the water molecules at the bottom basement better than the water molecules that are halfway to the basement, stuck between some carpets and flooring, because they’ accomplished their goal’.
Likewise if I came back in a time machine and told you that 100,000 years from now nearly every living human was descended in some way from your and that annoyingly odd birthmark you have on your butt is shared by everyone….well you may feel good or bad about that but it really doesn’t mean much for you. In 100,000 you ain’t going to be any less dead and buried.
It is a valid point that when language like ‘strategy’ and ‘goal’ is tossed around, it does make it sound like its some type of game, contest or competition which has some individualistic prize for ‘winning’. That, though, is just an artifact of the language and little else.
May 6th, 2011 | 6:16 pm
When choosing which brand of polygamist, why does the paper choose Mormons? Why not some Muslims (Osama and Obama’s fathers), why not some Native American tribes (Lakota Sitting Bull, Comanche Quanah Parker)? You couldn’t get published if this was about another group, someone sooner or later would lay down the race card.
May 6th, 2011 | 7:03 pm
TXW, I suspect it has more to do with available subjects and sufficient records — polygamist Muslims may exist in the U.S., but they’re even farther under the radar than polygamist Mormons. And I really don’t think we have good data on the families of 19th century native Americans. As far as I know polygamy has died out entirely among native Americans. Mormons are the best known and most visible polygamists in the U.S., and everyone knows they’re awesome at record-keeping.
Sometimes it’s just a matter of Occam’s Razor, you know.
May 8th, 2011 | 9:05 pm
Hi, Joe,
I guess I am puzzled about why you find this paper so objectionable. That wives in polygamous marriages produce fewer offspring the more wives there are sounds to me like a significant anthropological finding and an eminently publishable result. What do you find upsetting about it?
I am a little confused also about other aspects of your post. You quote with seeming approval the statement of David Stove to the effect that Darwinism predicts that every creature will produce as many offspring as is possible. Then you yourself argue that there may be reasons why limiting the number of one’s offspring might increase the number of one’s descendants in the long run, which if true refutes Stove’s assertion.
And I cannot make out whether you are saying that contraception reduces “evolutionary fitness” or are ridiculing that idea.
May 9th, 2011 | 2:21 am
I guess I am puzzled about why you find this paper so objectionable.
I was in a rush on Friday and didn’t take the time I should have to express what I objected to about this study. Basically, it seems, in my opinion, to fit into three categories that are on my list of pet peeves:
– Social science that masquerades as natural science.
–Precise claims about results that are predicated on imprecise assumptions and incomplete data.
–Results that are trivial or that do not prove what they claim to prove.
That wives in polygamous marriages produce fewer offspring the more wives there are sounds to me like a significant anthropological finding and an eminently publishable result. What do you find upsetting about it?
The main problem I have with it is that particular finding is neither new nor surprising. I suspect that no anthropological journal would have published that finding since it is both already known and in line with what we should expect to observe from real-world situations.
As for my claim that it is nothing new, Thomas G. Alexander wrote in his 1980 book The Mormon People: Their Character and Traditions, “Plural marriage, including monogamous remarriage, actually slightly suppressed the total number of children born.” Although there are likely a number of reasons why this was the case, I think we can show at least one reason why this result shouldn’t be surprising.
Utah in the late 1800s was a relatively rural state, so let’s compare two farmers who have comparable wealth (each has land equal to 20 acres) and are equally productive (each can produce enough to feed roughly one person per acre).
Farmer NP is a Mormon but not a polygamist (he’s the norm since only a third of LDS men had more than one wife). During his lifetime he has had 1 wife and 10 children. Farmer P is a Mormon polygamist. He has 3 wives (about the norm for polygamists) and 12 children.
Farmer NP has to feed 12 people (0.6 people per acre) while Farmer P has to feed 16 people (0.8 people per acre). Each of them could add a few more children to their brood without going hungry, but they are close the maximum number of children that most rural families would have at that time.
Now let’s consider this same scenario based on the assumptions of the evolutionary biologists who published this study.
Farmer NP’s wife had 10 children, a ratio of 10 to 1. Farmer P’s wives had 4 children each, a ratio of 4 to 1. What would happen if each of Farmer P’s wives had the same ratio of children as Farmer NP’s wife? Then Farmer NP would have 40 children (an increase of 28). The total number of mouths he would feed would now be 44—an average of 2.2 people per acre of farmland. The reason Farmer P doesn’t have the same ratio of children per wife as Farmer NP is obvious: If he did his children would starve.
Therein lies the problem with the paper’s assumption (which was derived from Darwin who got it from Malthus). The assumption is that animals (including humans) are expected to maximize the number of offspring they have. Of course this theory is—and always has been—disproved by observation of the facts. Environmental factors often limit the number of offspring—and that is just in the animal kingdom. For humans, the justifications for having fewer children than the biological maximum are legion. Not being able to feed them all, like the farmers in our example, is just one legitimate reason.
You quote with seeming approval the statement of David Stove to the effect that Darwinism predicts that every creature will produce as many offspring as is possible. Then you yourself argue that there may be reasons why limiting the number of one’s offspring might increase the number of one’s descendants in the long run, which if true refutes Stove’s assertion.
I’m not sure I understand how that could refute Stove’s assertion. Darwin’s view was that organic beings maximize the total number of offspring that is biologically possible. Having fewer offspring in order to increase the number of one’s descendants refutes Darwin’s claim, not Stove’s.
And I cannot make out whether you are saying that contraception reduces “evolutionary fitness” or are ridiculing that idea.
I don’t doubt the concept of “evolutionary fitness” has some explanatory value in evolutionary biology. But I think as it relates to this study (and specifically to humans) that it falls short of being useful. It seems that the researchers are using social science data and trying to shoehorn it into a natural science explanation. I believe there are a number of problems with this approach.
The first is with the significance and application, within this study, of the term “evolutionary fitness.” Relative fitness (which I believe is what the study is referring to) is defined as the average number of surviving progeny of a particular individual compared with average number of surviving progeny of competing individuals after a single generation. If polygamous wives have fewer offspring than non-polygamous wives then the former are, by this definition, less fit relative to the latter. It’s simply a tautology.
The second is that the results are unlikely to be replicable. The study presumes to be saying something about how polygamy relates to fitness. But if we did a comparative study based on, say, African polygamists, would the results be the same? Not likely. In fact, they are unlikely to be anywhere close since the number of wives and children in each polygamist situation varies considerably. The results of this study are based on a very specific culture (Mormons), in a very specific place (Utah), at a very specific time (the late 1800s). Assuming that it is more than a tautology (women who have more children have more children than women who have less children), then the findings should be universal to polygamist throughout the human species. If it doesn’t—and there is no evidence it does—then what’s the point?
The third issue I have is that the comparison is trivial. The claim is that the average polygamous Mormon woman in the late 1800s was less evolutionary fit than a non-polygamous Mormon woman. Okay, so what? If you compare that same group to almost any woman today they would be, comparatively speaking, more relatively fit. But why does the first claim matter at all? What advance of knowledge has been achieved by this finding? It doesn’t even tell us how the evolutionary fitness of these polygamists compares to other polygamists groups, either now or in the past so why should anyone care?
The study also passes over what could have been an interesting and significant finding. Because of the genealogical records, we could determine how many of the offspring of each of these women have survived up to modern times. If we were to find that fewer descendents of polygamous women were still around relative to non-polygamous women, then that would have been worthy of publication.
Of course the researchers don’t even bother to address that question because it would have required a lot of work. Instead, they take a relatively unreliable dataset (the 1910 census), fill in the gaps using questionable assumptions (their calculation of the number of infertile women is creative, but not reliable enough for an empirical study), and then presume to make natural science conclusions using speculative social science methods. That should have natural scientists crying foul.
And I cannot make out whether you are saying that contraception reduces “evolutionary fitness” or are ridiculing that idea.
My point about contraceptives was simply an a fortiori extension of the study’s finding. If polygamous women (Group A) are evolutionary less fit than non-polygamous women (Group B), then women today who use contraceptives (Group C) and have (statistically speaking) fewer children than Group A must also be less fit than Group B.
Although that’s a logical inference from the study, I suspect the researchers would be hesitant to draw that conclusion since it is politically incorrect to say that using contraceptives makes you less evolutionary fit than a Mormon sister wife in 1870.
May 9th, 2011 | 8:57 am
Joe Carter –
In Wikipedia terms… [citation needed].
May 9th, 2011 | 9:51 am
Ray Ingles In Wikipedia terms… [citation needed].
Origin of Species (Ch. 3):
May 9th, 2011 | 10:28 am
Joe,
I am still totally baffled. You do not link to the paper, but only to a news article that discusses the paper. From some of your comments it seems as though you are commenting on the research paper itself. So I looked up the research paper.
I just cannot see what has gotten you so exercised. The paper is a somewhat dry statistical analysis of the demographics of Mormons. You say that their conclusions (e.g. that wives in polygamous Mormon marriages have fewer children on average than those in monogamous marriages) are nothing new. What seems to be new is the size of the data set analyzed, which comprises tens of thousands of adults, who amongst them produced something like 600,000 children. If they have indeed done a more thorough statistical analysis based on a much larger set of data than had been done previously, then that is publishable research, even if the conclusions of the analysis do not contradict those of previous smaller studies. Publishable science is not just science that reaches new conclusions.
You make this statement: “Therein lies the problem with the paper’s assumption (which was derived from Darwin who got it from Malthus). The assumption is that animals (including humans) are expected to maximize the number of offspring they have.” I cannot find where this “assumption” is stated or used in the paper. The paper is almost entirely devoted to statistical analysis, and in particular calculating correlations among various quantities such as the “variance of reproductive success” and the “variance among mating success”. It is purely what is called “data analysis”. Only in the very brief “discussion” section at the end are there a few sentences on what all this may mean.
They do not make any grand metaphysical claims or grand claims to explain human behavior in evolutionary terms. It is basically demographics.
You say that their findings may have more to do with Mormon culture than anything more general and that a study of polygamy in Africa might give different results. Good point. But the only way to know is to do a careful study of the Mormon case and a careful study of the African case (or cases) and compare them. So they have done part of that task. Doubtless, at some point, some other people will do comparable studies of African polygamy. And that will get published. And then yet other researchers will compare these studies. That is how science works. What is your problem. You attack these guys for doing a study and then attack them for not having done additional studies.
Again, I am just totally baffled by your post and subsequent comments. This is a humdrum and dry-as-dust study, making modest claims about statistical correlations.
Does neo-Darwinian theory (and I don’t really give a rat’s tail what Mr. Darwin and Rev. Malthus may have thought, which is not very important except to historian) really imply that every individual organism maximizes the total number of offspring that is biologically possible? I think it is very much more complicated than that.
Is this possible that you may just be scratching your anti-darwinian itch in this instance more than responding to anything that these researchers really say? Just asking.
Steve
May 9th, 2011 | 11:45 am
Steve You do not link to the paper, but only to a news article that discusses the paper.
Sorry about that. Ray linked to the article in an earlier comment so I didn’t think to repost the link.
I just cannot see what has gotten you so exercised. The paper is a somewhat dry statistical analysis of the demographics of Mormons.
Indeed, and if this were a sociological analysis I wouldn’t have thought twice since it uses social science data and social science methods. But the conclusion purportedly says something about a natural science. Since that is the case, it should be held to the higher, moor rigorous standard that the natural sciences require.
What seems to be new is the size of the data set analyzed, which comprises tens of thousands of adults, who amongst them produced something like 600,000 children.
That’s not new either. They are simply using records from the Utah Population Database, which has been around for about thirty years. But that is the core of my problem with this study. The records that they are using are not nearly rigorous enough to be used to make claims about natural science.
If they have indeed done a more thorough statistical analysis based on a much larger set of data than had been done previously, then that is publishable research, even if the conclusions of the analysis do not contradict those of previous smaller studies.
I don’t disagree. If some researchers had simply run a statistical analysis on demographic data and published them in a journal on genealogy I’d have no objections. The readers of such a journal would recognize the inherent flaws in the data and would adjust their expectations accordingly.
But this paper takes a quite different approach. Because the data is not sufficient to do the analysis they require, they have to make all sorts of assumptions in order to get the parameters they need. They spend an entire section explaining how they account for the fact that the dataset doesn’t address nulliparous individuals (women who have never given birth). This is a huge hole in the data that can skew the results in a considerable ways.
If you can’t accurately count how many women are infertile or simply never had a child (and despite their hand-waving, they cannot) then you simply can’t draw an accurate conclusion about relative fitness.
It’s easy to show how the data can become skewed. Consider a married couple in which the husband or wife is infertile. Since they are not “parents” they do not show up in the dataset at all. On the other hand, a polygamous male will almost always show up since the church will not allow a man to take a second wife if he can’t sire children.
But what about the infertile polygamous wife? They are the ones that can really skew the data.
For the sake of simplicity, let’s reduce the dataset to two families—one polygamous and the other non-polygamous. Let’s assume that non-poly wife has had 5 children. The ratio of mothers to children on her side is 1:5.
In the polygamous family, the first and third wives each had 5 children but the second wife is infertile. What is the ratio on that side? 1:3.3.
So using the assumptions of this paper, we could determine that, on average, a polygamous wife is less relatively fit than a non-polygamous wife. Of course that is not reality, but it fits with the data.
“. . . The assumption is that animals (including humans) are expected to maximize the number of offspring they have.” I cannot find where this “assumption” is stated or used in the paper.
It is true that the paper doesn’t say that humans have to maximize their offspring in the way that Darwin meant (i.e., have as many babies as humanly possible). But their definition of “reproductive success” is based entirely on how many children a woman has with the assumption that she is more “evolutionarily fit” if she has more kids.
They do not make any grand metaphysical claims or grand claims to explain human behavior in evolutionary terms. It is basically demographics.
You are right that the paper is “basically demographics.” But if it makes no grand claims to explain human behavior in evolutionary terms then why is it published in a journal called “Evolution and Human Behavior.” Either the paper is considered to have some connection to explaining human behavior in evolutionary terms or it was published in the wrong journal (or possibly both).
That is how science works. What is your problem.
The problem I have is that they are doing social science work, not the work of evolutionary biology. If they had conducted a review of all demographic papers on polygamy and determined that there was a universalizable finding, then I would agree that they were on to something. But why was this published in a natural science journal when it is only applicable in one location at one specific point in time?
Imagine if a physicist were to publish a study that was determined to be applicable only in Texas in the 1930s, though further study might find that the finding is true everywhere and at all times. Would such a paper be published? I suspect it wouldn’t because physicist (I believe) have a higher standard for publication. Why then do we give biologist a pass on such incomplete findings?
Again, I am just totally baffled by your post and subsequent comments. This is a humdrum and dry-as-dust study, making modest claims about statistical correlations.
To reiterate, my point is that it is—at best—a social science paper that is based on flawed data and that is it is peculiar to publish it in a journal of natural science.
Is this possible that you may just be scratching your anti-darwinian itch in this instance more than responding to anything that these researchers really say? Just asking.
Although my problems with neo-Darwinism are legion (and tangentially related), this really isn’t about evolution at all. It is more similar to the issue I have with claims about global warming. The climate scientists also tend to use datasets that are not only incomplete, but that are considerably flawed (for example, there is no standardization in the way that surface temperatures are recorded and measured). When you start out with a questionable dataset, then the results should also be considered questionable. (For the record, I’m agnostic—or acognostic—about anthropomorphic climate change. I’m more apt to believe that it is true than I am to believe that we have the means to determine whether it is true or not.)
May 9th, 2011 | 12:04 pm
Joe, in one of those ellipses where you dropped some of Darwin’s text: “The only difference between organisms which annually produce eggs or seeds by the thousand, and those which produce extremely few, is, that the slow breeders would require a few more years to people, under favourable conditions, a whole district, let it be ever so large. The condor lays a couple of eggs and the ostrich a score, and yet in the same country the condor may be the more numerous of the two. The Fulmar petrel lays but one egg, yet it is believed to be the most numerous bird in the world. One fly deposits hundreds of eggs, and another, like the hippobosca, a single one. But this difference does not determine how many individuals of the two species can be supported in a district. A large number of eggs is of some importance to those species which depend on a fluctuating amount of food, for it allows them rapidly to increase in number. But the real importance of a large number of eggs or seeds is to make up for much destruction at some period of life; and this period in the great majority of cases is an early one.If an animal can in any way protect its own eggs or young, a small number may be produced, and yet the average stock be fully kept up…”
In any case, let’s assume you’re parsing Darwin correctly and completely. Darwin wasn’t a prophet laying down an unchangeable message, and indeed got many things wrong, like inheritance. Modern evolutionary theory recognizes many circumstances where organisms restrict their fertility to ensure ‘quality’ offspring; the main question is whether this happens before or after reproduction.
I reiterate my recommendation for David Sloan Wilson’s “Evolution For Everyone”; a lucid and plain-spoken introduction that addresses a lot of misconceptions (including the above) while giving a solid grounding on what evolution actually does propose.
May 9th, 2011 | 2:08 pm
Joe, Neither of us is in this field of research. Neither of us is competent to referee this paper — at least I am not. The paper was published in what I presume is a refereed journal. It is more than likely, therefore, that it was vetted by people who know something about the quality of the data they were used and the methods that were used to analyze it.
I too am skeptical of the claims of global warming alarmists. But that does not mean that I would be able to go into a journal of climatology, or geophysics, or atmospheric physics, etc and pick out a paper and make a meaningful judgment about its methods, conclusions, or publishability — except in extreme cases.
It sounds like you may have more background in social science research than I do — which wouldn’t be hard, since I have none. But do you really have enough background doing research in these areas that professional journals would send you papers of this sort to referee? If so, then I would defer to your judgment.
You say, “When you start out with a questionable dataset, then the results should also be considered questionable”. This is naive. People in the physical sciences often have to use data that is far from ideal. There are gaps, systematic errors, theoretical uncertainties, etc.
If I thought this was an important paper, making major scientific claims, or claims that had some relevance to philosophy, theology, or religion, I might spend the time to investigate it more closely or debate its merits. As it is, it seems a rather unremarkable, humdrum paper. I don’t think it illustrates any problem with darwinism, neo-darwinism, current scientific standards in the social sciences or natural sciences. I remain baffled by your interest in this paper.
My general approach is to leave technical disputes to the people expert in the relevant disciplines.
I will leave it at that. You can have the last word.
Links
Blogs
Find Us
Contact