Wednesday, May 30, 2012

Why Mothers Prefer Boys

Phil Kitcher is a philosopher who is interested in the philosophy of science and he's also very interested in evolution. In his recent article on The Trouble with Scientism he gives and example of .... well, I'm not exactly sure what.
The emphasis on generality inspires scientific imperialism, conjuring a vision of a completely unified future science, encapsulated in a “theory of everything.” Organisms are aggregates of cells, cells are dynamic molecular systems, the molecules are composed of atoms, which in their turn decompose into fermions and bosons (or maybe into quarks or even strings). From these facts it is tempting to infer that all phenomena—including human actions and interaction—can “in principle” be understood ultimately in the language of physics, although for the moment we might settle for biology or neuroscience. This is a great temptation. We should resist it. Even if a process is constituted by the movements of a large number of constituent parts, this does not mean that it can be adequately explained by tracing those motions.

A tale from the history of human biology brings out the point. John Arbuthnot, an eighteenth-century British physician, noted a fact that greatly surprised him. Studying the registry of births in London between 1629 and 1710, he found that all of the years he reviewed showed a preponderance of male births: in his terms, each year was a “male year.” If you were a mad devotee of mechanistic analysis, you might think of explaining this—“in principle”—by tracing the motions of individual cells, first sperm and eggs, then parts of growing embryos, and showing how the maleness of each year was produced. But there is a better explanation, one that shows the record to be no accident. Evolutionary theory predicts that for many, but not all, species, the equilibrium sex-ratio will be 1:1 at sexual maturity. If it deviates, natural selection will favor the underrepresented sex: if boys are less common, invest in sons and you are likely to have more grandchildren. This means that if one sex is more likely to die before reaching reproductive age, more of that sex will have to be produced to start with. Since human males are the weaker sex—that is, they are more likely to die between birth and puberty—reproduction is biased in their favor.
In humans, the average sex ratio at birth is about 105 boys to every 100 girls but this ratio varies a lot from country to country and it depends on environmental conditions. There are many factors that affect fertilization and the survival of embryos and fetuses.

Is it reasonable to believe that the observed sex ratio (1.05) is the product of natural selection? You can't really answer that question until you know the mechanism of altered sex ratios. What is being selected? Is it the probability that a male sperm will reach the egg before a female sperm? If so, what kind of selective advantage would have to apply to change that probability from from 50% to 51% or 52%? How is it done? What alleles are involved?

Why does Philip Kitcher, a philosopher of science, think that a postulated adaptive explanation is a "better explanation" than a mechanistic one? Don't you actually have to "prove" your adaptive model at the level of genes, cells, and developing embryos before it can be accepted?


29 comments :

  1. You can't really answer that question until you know the mechanism of altered sex ratios

    Maybe, because the X chromosome is larger than the Y, the female-producing sperm are slower swimmers ;-)

    I jest, but the explanation I pulled out of my ass appears to have greater explanatory power than Kitchers selection-ism

    ReplyDelete
    Replies
    1. It's a very common explanation. I agree that it seems superior to Kitcher's explanation. I'm having trouble understanding his point.

      Delete
    2. It's a very common explanation
      It is? I just made it up...

      ...I don't suppose there is any data actually supporting that? I'm far from a sperm expert, but the mass difference of an X vs Y must be quite small in relation to the total mass of a sperm. If this is true, colour me surprised!

      Delete
  2. Suppose we analyse sex-ratios for a number of different species and find that they all have a sex-ratio of 1:1 at sexual maturity, but have different sex-ratios at birth. In species where more males are killed before maturity, the sex ratio at birth is skewed more heavily to males,and so on. In each case, the greater number of males at birth is exactly cancelled out by maturity. In this case, I would suggest that we have good evidence for the adaptive explanation, even if we have no knowledge of any mechanism that underlies the adaptation. Moreover, even if we did know the mechanism at the genetic level, there is no reason to suppose that the same mechanism would be involved in different species. And if there are multiple different mechanisms, all achieving the same result, the existence of such mechanisms is best explained in terms of their common adaptive role.

    I don't know if this is the picture that Kitcher has in mind, but I suspect it is, given his unificationist account of scientific explanation.

    ReplyDelete
  3. My response is along the line of Iain's. For decades, scientists have been investigating selection on sex ratios. These studies have included theoretical work, as well as observations and experiments. For example, theory predicts that under some conditions, selection should favor the ability to adjust sex ratio for maximum advantage in different environments. There's lovely work on the Seychelles warbler (to pick just one species) that shows how tipping in favor of males versus females meets the prediction of the hypothesis: for example-- http://onlinelibrary.wiley.com/doi/10.1046/j.1461-0248.1998.00009.x/abstract

    Note that the authors of that paper did not work out the proximate cause of the sex ratio manipulation. Instead, they tested an adaptive scenario.

    This is not to say that one approach is better than the other. They can be complimentary, just as Nikko Tinbergen proposed decades ago: http://en.wikipedia.org/wiki/Tinbergen's_four_questions

    ReplyDelete
    Replies
    1. My response is along the line of Iain's. For decades, scientists have been investigating selection on sex ratios.

      I agree that for decades scientist have been assuming that there's an adaptive explanation for sex ratios that deviate from 1:1. I agree that they have been looking for any evidence to support their adaptive explanations. I also agree that after all those decades the work on the Seychelles warbler (1994?) keeps coming up time and time again. It's sort of like trotting out sickle cell anemia whenever you want to prove that balancing selection is important.

      I think we can also agree that after decades of work, nobody has a mechanistic explanation. I don't even think that anyone knows which "sex ratio" genes are involved in this "adaptive" response.

      Carl, please give me your best explanation for why the birth ratio in humans is about 1.05 in favor of boys. Any supporting evidence would be a major bonus.

      I don't think we can eliminate the possibility that the human sex ratio has no adaptive significance. Maybe it's just because male sperm swim faster.

      Delete
    2. I will note that, taking the philosopher's prerogative, my argument was hypothetical: to establish that an adaptive account *could* be explanatory in a way that no mechanistic account could match, it is only necessary that my suppositions be *possible*, not that they are *true* (truth is for you biologists to decide!). So Kitcher's philosophical point could stand, even if his example is bad biology.

      Delete
    3. Larry, are you saying that the Seychelles warbler is the only example of an extreme sex ratio for which there is experimental evidence for adaptation? That's not true. I could list others. For example, fig wasps. http://rspb.royalsocietypublishing.org/content/275/1643/1603.abstract

      I didn't catalog the other examples because this is just a comment post.

      These examples provide evidence that deviations from a 1 to 1 sex ratio can be favored by selection. That's the only point I was trying to make. Is the 1.05-1 ratio in humans adaptive? It's a reasonable question to ask, but the answer might well be no. Maybe there is no adaptive significance, as you suggest. I don't like Kitcher's take on this, because he simply assumes there must be an adaptive explanation, without even trying to offer evidence to reject the non-adaptive alternative. (Maybe I could have made that clearer in my first comment.)

      That being said, there are scientists who are testing adaptive explanations for human sex ratio. Here's one example that just came out: http://humrep.oxfordjournals.org/content/early/2012/02/01/humrep.der480.full

      As for a mechanistic explanation, there's plenty of work on birds, although it is quite a murky subject. PDF: http://www.u.arizona.edu/~abadyaev/pubs/107.pdf

      Delete
    4. Carl asks,

      Larry, are you saying that the Seychelles warbler is the only example of an extreme sex ratio for which there is experimental evidence for adaptation?

      No, I'm not saying that there's nothing in the scientific literature to support the hypothesis.

      One the other hand, I'm skeptical of many of these studies since the results don't make a lot of sense to me and since I'm generally suspicious of overly adaptive explanations.

      As a general rule, I do not assume that every paper published n the scientific literature is necessarily correct. As a general rule, when advocating an important evolutionary mechanism, as this is, I expect there to be abundant evidence from a wide variety of species and a great deal of reproducibility and confirmation of published results.

      Delete
    5. It's sort of like trotting out sickle cell anemia whenever you want to prove that balancing selection is important.

      Then update yourself by reading one of the leading experts in the field.

      "Balancing Selection Maintains a Form of ERAP2 that Undergoes Nonsense-Mediated Decay and Affects Antigen Presentation" PLoS Genet. 2010.

      "Targets of Balancing Selection in the Human Genome" Molecular Biology and Evolution. 2009.

      Delete
    6. Then update yourself by reading one of the leading experts in the field.

      I think you missed the point. Scientific papers in apparent support of important concepts are a dime a dozen. What really counts in the long run is whether the data survives years of close scrutiny and attempts to refute it. In the case of balancing selection, after sevetal decades of work we only have a few solid examples that have stood the test of time.

      That's enough to establish that the phenomenon exists but not enough to establish that it's common. In fact, it seems to be quite rare.

      Delete
    7. Your original point was a complaint about trotting one tired example to prove the importance of a mechanism of natural selection. I simply gave you another example. You now have two recent papers, one of which involves the major histocompatibility complex, "arguably the best-established target of natural selection in vertebrates." Again, your original complaint was about balancing selection's importance and not how commonly it occurs (which, by the way, no one is debating so it's irrelevant). I hope you can appreciate the non-equivalence of those two words you casually exchange in your future grumblings on the subject.

      Delete
  4. The argument is stupid independent of the unsupported claim of adaptiveness. Kitcher wants to illustrate why ultra-reductive approaches are flawed, but his dichotomy is false. First of all, if you wanted to understand the mechanism of biased sex ratios at birth, you most certainly would look at all the parts involved. It's a totally different question from how the bias became established in humans, which is potentially an evolutionary question.

    And in any case, just because there are things we can't currently understand at the level of molecules doesn't mean such an understanding is impossible!

    ReplyDelete
    Replies
    1. You are missing Kitcher's point. He certainly doesn't deny that there are mechanistic stories to be told, even if we don't know them now. Nor does he deny that the reductive account is explanatory. What he claims is that, even if you have the mechanistic account, that account by itself misses out an important piece of an adequate explanation of the phenomena. And, sometimes, the *best* explanation of the phenomena is a non-reductive one.

      Delete
    2. I disagree. Kitcher claims the evolutionary explanation is better than the molecular/cellular one. But it's not, because the two approaches are addressing different questions. If you want to know how Homo sapiens came to exhibit sex bias as a species, the evolutionary explanation is indeed best. But if you want to know the 'mechanics' of how the bias occurs, the molecular/cellular explanation is best.

      And again, even if there are things we can't currently understand at an ultra-reductionist level, that's hardly proof that it's not possible in principle. Kitcher believes that, and he may even be correct, but his argument and his illustration do not support his belief.

      Delete
    3. The ultra-reductionist holds that, if we have a reductive explanations of some phenomenon, any non-reductive explanation of the phenomenon is irrelevant - it adds nothing to our understanding. Kitcher provides a case in which, arguably, the evolutionary (non-reductive) explanation of how humans came to exhibit sex-bias as a species is the best explanation of that phenomenon, even if we had a molecular/cellular (reductive) explanation. If a case like this is possible, then the ultra-reductionist is wrong. You seem to agree that the evolutionary explanation is best, regarding this question, which is all Kitcher needs for his argument to go through. Pointing out that the molecular explanation is better in answer a *different* question is entirely irrelevant.

      And again, Kitcher doesn't deny that it is possible to understand phenomena at an ultra-reductionist level, he just denies that, in this sort of case, that understanding would provide a *better* explanation than the nonreductive one. In my first comment, I provided an argument for why this might be so.

      Delete
    4. Iain says,

      Kitcher provides a case in which, arguably, the evolutionary (non-reductive) explanation of how humans came to exhibit sex-bias as a species is the best explanation of that phenomenon, even if we had a molecular/cellular (reductive) explanation.

      Yes, that's exactly what Kitcher is saying. It's a very silly thing to say.

      Imagine that we had a mechanistic explanation of the sex ratio. Imagine that it had to do with the fact that male sperm swim a bit faster than female sperm because the Y chromosome is so much smaller than the X chromosome.

      Would you still think that the adaptive evolutionary explanation was the best explanation?

      The adaptive evolutionary explanation is nothing more than speculation. It can never, ever, trump facts.

      Delete
    5. Imagine that we had a mechanistic explanation of the sex ratio. Imagine that it had to do with the fact that male sperm swim a bit faster than female sperm because the Y chromosome is so much smaller than the X chromosome.

      This would presumably be a scenario in which the adaptive explanation was, in fact, false. That it was false would presumably show up in other species, since the sex ratios at birth would vary with the relative sizes of X and Y chromosomes, which I assume would not be correlated with male survival rates.

      In this case, it is obvious that the (true) mechanistic explanation is better than the (false) adaptive one. And perhaps that is the correct thing to say about human sex ratios - I know nothing about the topic, but from what you say, it sounds like the evidence does not favour the hypothesis Kitcher advances. But this isn't an interesting case - it is trivial that a true explanation is to be preferred to a false one.

      The interesting case is where both reductive and non-reductive explanations are true - or, at least, consistent with all evidence. So suppose the evidence was actually consistent with the adaptive explanation. Suppose that mechanisms were found that did, in fact, balance out sex ratios at maturity by varying the ratios at birth - and did so consistently across species. In that case, I think Kitcher would be right in claiming the adaptive explanation would be the better explanation, because it shows that the balancing out is not merely coincidental - something the purely mechanistic explanation leaves out.

      The adaptive evolutionary explanation is nothing more than speculation. It can never, ever, trump facts.

      In my first comment, I tried to provide a scenario in which the adaptive evolutionary explanation was well supported by evidence. Would you accept this explanation if the evidence were as I outlined?

      It is easy to be distracted by the fact that Kitcher uses what he takes to be a true instance of the scenario he envisions, and to ask whether it is, in fact, true. To make his philosophical point, Kitcher could have just as easily used an imaginary example.

      Delete
    6. Iain,

      Kitcher's thesis is that some things can never be fully explained at a reductionist level. You can't support that claim by comparing high-level and low-level answers to different questions. You need to show that a high-level answer to some question is better than the best low-level answer to the same question.

      More importantly, you have to show that the high-level answer is better than any possible low-level answer.

      Yes, I agree the high-level approach is the best for understanding how sex bias evolves in a species, for exactly the reasons you illustrated in your first post. Our current knowledge isn't sufficient for a reductionist approach to that question.

      But that wasn't Kirtcher's claim. His claim is that reductionist answers to some questions are inferior even in principle. Perhaps that's true, but Kirtcher failed to give a cogent argument for that claim.

      Delete
  5. I'll hypothesize in the other direction, that female embryos are disfavored due to improper X-inactivation causing some embryonic lethality from the XX zygotes.

    ReplyDelete
    Replies
    1. Then the phenomen should occur in all Eutheria. Does it?

      Delete
    2. And wouldn't it be the opposite in Birds, Makes are ZZ and Females are ZW.

      Delete
  6. Why does Philip Kitcher, a philosopher of science, think that a postulated adaptive explanation is a "better explanation" than a mechanistic one?

    It seems that some people prefer a teleological explanation to mechanistic explanation. For that matter, at its core, the creation/evolution debate is that creationists want a teleological explanation while scientists want a more mechanistic explanation.

    Don't you actually have to "prove" your adaptive model at the level of genes, cells, and developing embryos before it can be accepted?

    Yes, that's very much the point. Where there is an apparent teleological explanation, scientists will want to dig further and find the underlying mechanisms that make it work. Those who prefer teleological explanations don't seem to understand this.

    Perhaps this is at the heart of "The Two Cultures" division, and possibly even the core issue that drives creationist thinking.

    ReplyDelete
  7. In the piece, it seems that Kitcher's main focus is simply to illustrate the anti-reductionist point in contemporary philosophy of science (which some suggest is the consensus view).

    A lower-level reductive account of a computer's electrons, while certainly a requisite proximate component of the full mechanistic explanation of the software being subserved by it, tells you nothing about the higher-level facts of the software itself. That is, if you stripped down Microsoft Word to the level of its electrons, you would know nothing about the program itself qua word processor.

    Unfortunately, those unfamiliar with the apposite philosophy of science literature will be reflexively off-put by the term 'anti-reductionist'.

    The anti-reductionist consensus is not to diminish the importance of more reductive accounts of a given explanandum; it's merely to highlight that those accounts aren't the full story. Indeed, if you really want to gain a good understanding of Microsoft Word -- or any other application for that matter -- you are best to eschew the physical explanation altogether and focus your explanatory probings at the higher level; much like how dabbling at the level of molecules is a poor strategy for attempting to disinter the teleonomy (the evolutionary rationale) of the eye.

    ReplyDelete
  8. N spake thusly:"That is, if you stripped down Microsoft Word to the level of its electrons, you would know nothing about the program itself qua word processor."
    If you could track the output of MSWord with the 'base' level action of electrons, then wouldn't you actually have a better understanding of it? If you wanted a word to appear or a function to be applied, you could change it using 'electron-speak', argueably you'd have a better understanding of the system than if you were a programmer for MS. Would you argue that a person that can recreate MSWord from something like Machine Language has a better understanding of it than a person that can only recreate it in a 'higher level' programming language?

    ReplyDelete
  9. I feel like Prof. Kitchner's point is that the 'higher level' explanation is equal, if not better, than the lower level explanation.
    But, in the evolutionary example, isn't this a bit like saying that adaptations occur for the benefit of a species/population/individual? Whereas in reality there is merely variation and elimination? If we have an adaptive explanation, isn't the next step to find out why/how the adaptive "explanation" occurs?

    Is there an analogy for this in the social sciences? If a Freudian or Marxist interpretation/explanation is arrived at, how can it be challenged? My understanding is that they can't be internally checked, a Marxist criticism can't show us if the Marxist interpretation is correct, and Marxism and Freudianism can't talk to each other, but Science can show if/when both are wrong, and even can show us when science is wrong. Thus science came up with phlogiston and aether, but also did away with them, whereas dialetical materialism isn't refered to much in Marxist circles today /merely/ because it has fallen out of popularity as a fad (or, one could argue, been refuted by science).

    ReplyDelete
    Replies
    1. "If you could track the output of MSWord with the 'base' level action of electrons, then wouldn't you actually have a better understanding of it?"

      If one avails themselves of the 'vocabulary' at merely the reduced ontological scale of electrons, one would not 'see' the functional vocabulary which 'emerges' (in a completely physical, non-spookily manner, I might add) at the higher ontological scale(s) wherein the program is most aptly described.

      Your point regarding machine language is notable; however it's important to keep in mind that machine language is still not equivalent to the ontological level found in the electron illustration, and thus is rather disanalagous. (That is, electrons stand in relation to Microsoft Word differently than the relation machine language stands to Word.)

      The electron example is useful because it presents a more maximally contrastive illustration for conveying the anti-reductionist philosophy of science.



      "But, in the evolutionary example, isn't this a bit like saying that adaptations occur for the benefit of a species/population/individual? Whereas in reality there is merely variation and elimination? If we have an adaptive explanation, isn't the next step to find out why/how the adaptive "explanation" occurs?"

      There is of course variation and elimination. This is certainly a proximal description of what occurs; though there is much more to an adaptive episode in evolutionary history than just such a proximate/mechanistic descriptive account. For a more complete explanation, one also wants to know why a given adaptation has emerged -- therefore one must ascend up the ontological hierarchy to gain purchase on the requisite explanatory details.

      Delete
    2. "If one avails themselves of the 'vocabulary' at merely the reduced ontological scale of electrons, one would not 'see' the functional vocabulary which 'emerges' (in a completely physical, non-spookily manner, I might add) at the higher ontological scale(s) wherein the program is most aptly described."

      Well, sure. But that's because the vocabulary at the higher ontological scale is dependent on something external to the behavior of the electrons - human language. Following the behavior of the electrons can provide a deep understanding of how a keystroke causes the appearance of specific pixel patterns (i.e. letters) on a screen. It can't provide understanding of what those patterns mean because that's not a function of the the electron's behavior. It's not even a function of the MSWord program itself. It's a function of the human user who assigns meaning to the patterns that appear on-screen.

      This is similar to my previous objection. You're essentially trying to compare high- and low-level answers to different questions.

      Delete
  10. I looked this subject up because my Mom favors the 5 boys over the 2 girls in my family.

    I was thinking it was because she (my mother) came from a family of 5 girls and her Mother (my grandmother) came from a family of 5 girls. I know my Grandmother always went gaga over her grandson's.

    ReplyDelete