This reflects cultural conventions, not sexism

Wait wait wait.

Yet another of the American Enterprise Institute’s anti-feminist “feminists” takes on “the conventional wisdom” about women in science.

It’s the conventional wisdom that women are held back in science because of sexism. A new paper by a research team at Cornell University reports that young women faculty members prosper in math-based fields of science. Statistically, women are less likely to continue on in certain science fields, but there are cultural conventions that need to be taken into account. Visiting Factual Feminist Sally Satel will discuss these factors in this episode.

But there are cultural conventions? Well of course there are – and those cultural conventions are part of what sexism is. Why would the presence of cultural conventions undermine claims that women are held back in science because of sexism?

But Satel says it does.

She says women and men without children have pretty equal rates of publication, but men with children have higher rates than women with children. But don’t go thinking that’s sexism – oh no, it’s something completely different.

More research is needed to fill in this picture, but you could speculate that this disparity exists because fathers are more likely to have a spouse who’s caring full time for the children than are the mothers. It’s not rocket science. It’s easier to have kids when there’s someone at home doing the childcare. This reflects cultural conventions, not sexism in science. [1:50 to 2:20]

I haven’t yet listened to how she explains that absurd claim, because I was so gobsmacked by it that I wanted to yell about it first.

Hello? The idea that women must or should do all the childcare while men do none of it (or share it only in leisure hours) is indeed a cultural convention, and that cultural convention is sexist. Defining people’s duties and tasks by what gender they are is indeed sexism. The fact that men are more likely to have a spouse who’s caring full time for the children than are women is indeed sexism.

Let me guess – Satel’s explanation will be that this pattern simply reflects people’s natural unforced preferences, now that we have totally eradicated sexism and thus made all preferences free.

Ha. Like hell we have.


  1. Athywren; Kitty Wrangler says

    Yeah! It’s not cultural sexism, it’s cultural convention! Way different spelling. Therefore it’s cool. Woohoo.

  2. doublereed says

    Rate of publication is such a specific statistic, it really makes me think there is some major cherry-picking going on here (even with that blatantly obvious error you point out). I know she said some various other statistics which showed progress, but I think I’d have to read more in-depth.

    Many of the other statistics seemed very specific like “tenure-track associate professors” which, from my understanding of academia, is quite a small amount. Many universities have become far stingier with tenure than ever before. Makes me highly skeptical.

  3. josefjohann says

    This reminds me of common arguments you hear against the existence of the wage gap. “There’s no wage gap because women leave work for childcare.”

    Right, but that’s just part of what the wage gap is. It’s not so much disputing the empirical phenomenon, it’s just slicing off certain parts of that phenomenon and claiming they “don’t count” for various reasons.

  4. Hj Hornbeck says

    Never heard of the name before, so I went on a quick hunt to find out more about Sally Satel. I quickly found this:

    In practicing medicine, I am not colorblind. I always take note of my patient’s race. So do many of my colleagues. We do it because certain diseases and treatment responses cluster by ethnicity. Recognizing these patterns can help us diagnose disease more efficiently and prescribe medications more effectively. When it comes to practicing medicine, stereotyping often works.

    Yep, she sounds pretty racist. As a reviewer to one of her books put it:

    It is clear that the authors recognize that inequalities exist in health outcomes among different groups in this country. They also identify several significant contributing factors to the health disparities gap. What is not clear is why they feel the need to try and dispel the existence of racism in this country; try to recharacterize the expression of racism in different, more palatable terms; attribute the socioeconomic effects due to racism to other causes; and reframe the discussion in terms of class rather than race.

  5. Hj Hornbeck says

    And would it shock you to learn both Satel and Sommers are very skeptical of PTSD?

    PTSD did not formally exist as a diagnosis until 1980, when the American Psychiatric Association added it to its manual. Fear reactions are ages old. They were recognized in the Civil War; in World War I it was called shell shock; in World War II, battle fatigue. But when the diagnosis was advocated in the mid ’70s by self-proclaimed anti-war psychiatrists, it took on new aspects. For the first time, PTSD was considered a normal response to stress. It was also believed to have delayed onset. A person could supposedly function well for months or years after a traumatic experience and then suddenly become irritable, possibly violent, unable to hold a job, depressed. PTSD turns out to be not as common as many mental health professionals suggest. And it is often colored by political considerations.

    Right after 9/11, self-appointed trauma counselors flooded Ground Zero, trying to engage New Yorkers, asking how they felt. This was an astounding waste of resources. What we know historically from responses to disasters is that most people handle these things pretty well on their own. It makes no sense to intervene within the first few weeks, because intense emotional responses then are normal. If there is a pathology, it is not usually detectable until at least a month later. In some subset of recipients early intervention actually impairs recovery.

  6. Al Dente says

    And would it shock you to learn both Satel and Sommers are very skeptical of PTSD?

    The Roman Army recognized PTSD and developed reasonably effective ways of dealing with it. Basically senior NCOs would counsel the affected legionnaires, being sympathetic and helping the solder deal with his stress. The legionnaires would be kept in the combat zone and remain under military discipline. The Romans knew that some soldiers didn’t recover from PTSD and, after examination by medical personnel, would be medically discharged.

  7. themann1086 says

    Me thinking to myself: “My opinion of Sommers cannot possibly get any low-”

    And would it shock you to learn both Satel and Sommers are very skeptical of PTSD?


  8. Pierce R. Butler says

    Visiting Factual Feminist Sally Satel …

    Is “Factual Feminist” a regular position in the AEI, with Resident and Visiting job slots?

    Do they get paid 3/4 of what their equivalent Fellows receive?

  9. karmacat says

    I just found out she is a psychiatrist so she should know better when talking about PTSD. But I guess that’s what happens when you sell your soul to the American Enterprise Institute. People are not given the diagnosis of PTSD unless their symptoms persist after 4 weeks. Before then it is called Acute Stress disorder. But early intervention is still very useful and may prevent some symptoms of PTSD. the diagnosis of PTSD is always being researched and refined over the decades.

    In 1980, DSM 3 came out and was very different from the first two. It was a more detailed list of diagnoses and symptoms with the purpose of using this for research. If everyone is on the same page in terms of diagnosis, then the research is more useful.

    Satel complains about PC taking over medicine, but it is important. Research has shown that blacks are more often diagnosed with schizophrenia than bipolar yet more whites are more likely to be diagnosed with bipolar disorder. This obviously shows an “unconscious” bias when diagnosing different patients. I am now aware of this bias so hopefully I can come up with an accurate diagnosis for a patient.

    there is more I could say, but clearly Satel if full of crap, to put it delicately

  10. Jeremy Shaffer says

    Maybe I’m just not reading it right but does this part from the quote in HJ Hornbeck’s #5:

    A person could supposedly function well for months or years after a traumatic experience and then suddenly become irritable, possibly violent, unable to hold a job, depressed.

    seem contradictory with this:

    It makes no sense to intervene within the first few weeks, because intense emotional responses then are normal.

    In the first part they seem think a laughable aspect of PTSD diagnosis is that it might be months later before it manifests, yet turn around and say you shouldn’t expect to see it right away either. Like I said, i may just be reading this wrong but it seems they are just saying what ever they feel is needed in order to paint PTSD diagnosis in a negative light.

  11. GroceryDancer says

    @HJ Hornbeck (4):

    I’m ethnically Jewish. This means that I’m more at risk for certain specific genetic conditions (such as Tay-Sachs, to name the most well-known). If I go in for testing, I want the doctor to be taking that into account and checking me for things that I’m particularly susceptible to, even if the incident of such conditions averaged across the whole population is small enough that it wouldn’t ordinarily make sense to test for that.

    Now, if the doctor said, “You’re Jewish, so that explains your long nose,” then yeah, I’d be pissed off. But I don’t think Satel is saying that.

  12. Robert B. says

    Isn’t “cultural conventions that hold women back” the definition of sexism? Of what else does sexism consist?

  13. says

    A generous interpretation could be that she is claiming that the scientific community, itself, isn’t the source of the problem, but the broader cultural sexism of women being expected to be the primary child-rearers is. This may be a major contributing cause and certainly needs to be addressed in our society, but that just means that those in the STEM fields need to take that into account and not just wash their hands of it and say “not our problem.” Other fields manage it without much issue, so it seems they should be able to do something.

  14. Pteryxx says

    re @12 and @13 –

    Yes, that – I didn’t see that thing Satel said as racist either, for the same reason. Just sensible statistically-based practice I would think.

    It would be, if the statistics hadn’t been skewed by racism in the first place, when they were established. One famous example is the “race setting” on spirometers, which measure lung function.

    How Racism Creeps Into Medicine

    The notion that people of color have a racially defined deficiency isn’t new. The 19th century practice of measuring skulls, and equating them with morality and intelligence, is perhaps the most infamous example. But race-based measurements still persist. Today, doctors examine our lungs using spirometers that are “race corrected.” Normal values for lung health are reduced for patients that doctors identify as black. Not only might this practice mask economic or environmental explanations for lower lung capacity, but the logic of innate, racial difference is built into things like disability estimates, pre-employment physicals, and clinical diagnoses that rely on the spirometer. Race has become a biologically distinct, scientifically valid category despite the unnatural and social process of its creation.


    Hamza Shaban: How did the idea of race corrections and differing lung capacity come about?

    Lundy Braun: My research suggests that Samuel Cartwright, a Southern physician and plantation owner, was the first person to use the spirometer to compare lung capacity in blacks and whites. The first major study making racial comparisons of lung capacity with a large sample size was the anthropometric study of Union soldiers directed by Benjamin Apthorp Gould, published in 1869.

    The idea about the pathology of black lungs circulated in medical groups in the late 19th century but the next scientifically modern racial comparison was published in the Journal of the American Medical Association in 1922. This paper was followed by a flurry of studies in the 1920s, some of which continue to be cited in the 2000s. Gould’s book also continues to be cited.

    Shaban: So within the medical community this is a well-established concept?

    Braun: If you look at the scientific literature, virtually everyone in the world has lower lung capacity than people classified as whites. There is a scientific consensus. The question I’m interested in is: How did this idea of difference get into science? And how was difference explained? The problem here is the survival of the framework of innate racial difference.

    Shaban: Race correction is actually built into the spirometer, right?

    Braun: When I interviewed physicians they were sort of vaguely aware of race correction. But they don’t necessarily know that they’re activating a correction factor when they push the button or select a certain drop-down menu. Some even argued that they didn’t race correct, interestingly enough, but when I looked at the specification sheet, a correction factor was built into the machine.

    Because of the race setting, black patients with lower lung capacities get marked as in the “normal” range, meaning they have to be sicker to qualify for disability, among other things. This also camouflages the detrimental effects of pollution and poverty.

  15. Pteryxx says

    Another couple of examples of race bias built into medicine:

    Race is engrained in American medical practice. Sometimes beliefs about racial difference are even wired into medical diagnostic machines. For example, you can’t get a bone scan evaluated without designating a race, because the formulae programmed into bone densitometers use different standards for assessing bone thinning in white, Asian, Hispanic, and African American women. The evidence supporting different standards is rarely questioned and certainly unknown to the technicians who operate the machines. Often even the radiologists who evaluate the results don’t know much about the differing standards.


    In Chicago in 1980, black and white women died of breast cancer at the same rate. Today, despite being slightly more likely to get breast cancer, white female Chicagoans are half as likely to die from it. Could the difference in death rates be due to genetic differences between black and white women? A wealth of evidence suggests otherwise.


    Which brings us back to breast cancer. Not only rates, but breast cancer patterns differ between black and white women. When diagnosed, black women are more likely to be under the age of 35 and to die by the age of 50. Some have argued that their tumors spread more quickly because they differ physiologically from white women. Black women tend to lack key hormone receptors, which means that tumors respond poorly to familiar hormone-based treatments.

    Physician and cancer researcher Olufunmilayo Olopade noticed these differences and originally assumed that they were due to genetic, race-based differences between white women and women of African origin. Recently, however, she has begun to see things differently, looking to how women of color embody the daily stresses of racism and economic deprivation.

    Boston Review – Bodies with Histories

  16. Pteryxx says

    One more article interviewing Braun:

    There were 94 groups defined as racial or ethnic that came up in our systematic review, whose lung function was compared to whites. (Researchers used a variety of terms such as Western or Caucasian to define whites.) The vast majority of studies found that every group in the world, with few exceptions, has lower lung capacity than whites. That in itself raises certain interesting questions. Of those that did find lower lung capacity, a majority of papers explained difference as either innate or due to anthropometric factors with about 23 percent citing environmental factors.

    So despite the fact that this central variable isn’t defined, a significant number of researchers explained difference as fixed and then on this basis corrected for race.

    Only 6 percent of the papers looked at socioeconomic factors. We were surprised by the degree to which socioeconomic status and race were disconnected.


    Increasingly there is a turn to what is termed race-based medicine. We see race-based medicine in many contexts, including understandings of disparities in blood pressure and heart disease. Glomerular filtration rates are reported according to race. There have also been debates around growth curves.

    There was a major controversy widely reported in the press in the 2000s around a drug called BiDil developed to treat congestive heart failure. BiDil comprises two widely available generic drugs. After the researchers failed to get patent approval they turned to race, making poorly supported arguments to the FDA for its unique benefits in African Americans.

    >How do physicians define race if they are going to be applying differential standards?

    Rather than looking at social factors, some physicians explain that higher rates of disease are due to genetics. One example is what’s referred to as the salt-slavery hypothesis—it’s a hypothesis about the cause of high blood pressure in African Americans that has been completely debunked but maintains credibility in the clinic.

    Anyway, there’s a bunch of evidence for unexamined “innate differences” built into the assumptions underlying medicine. People really, really, really want to explain systemic bias away as biology.

  17. karmacat says

    Thanks Ptyrexx for the info. People judge race based on external characteristics but that only tells part of an individual’s genetic make-up. I have a patient whose diagnosis of Crohn’s was delayed because her doctors didn’t think blacks got Crohn’s.

    When I was a medical student I had a VA patient who needed home oxygen but his number wasn’t low enough. So I was instructed to have him walk up stairs until his number was low enough. Sometimes the system can prevent patients from getting the best care. Most doctors are white and there is an unconscious bias to want to help someone who looks like you. Not being aware of these biases can negatively affect how patients are treated

  18. johnthedrunkard says

    All that obsessive skull measuring missed the fact that the only correlate for skull size is BODY size. I’m just remembering from Gould’s ‘Mismeasure of Man,’ but I think the argument holds up.

    G.B. Shaw wrote that one can use statistics to prove that wearing a top hat is a cure for malnutrition. There ARE, essentially trivial, differences between sub-Saharan Africans and Northern Europeans. I’d never heard the claim about lung capacity before. Does it hold up as well as the debunked claims about skull volume?

    And how about the ‘cultural convention’ by which black skin exerts a magnetic attraction for bullets?

  19. Hj Hornbeck says

    Pteryxx did an excellent job of pointing out some of the racism in medicine. But there’s a greater problem here: what is a “black” person? In some countries, Barack Obama wouldn’t qualify and even in the USA Americans aren’t sure what race he is. So if the US president waltzed into a walk-in clinic for depression, his starting level of Prozac would depend more on the opinion of the doctor he sees than his own physiology.

    These lines in Dr. Satel’s article raised huge red flags for me, too:

    Almost every day at the Washington drug clinic where I work as a psychiatrist, race plays a useful diagnostic role. When I prescribe Prozac to a patient who is African-American, I start at a lower dose, 5 or 10 milligrams instead of the usual 10-to-20 milligram dose. I do this in part because clinical experience and pharmacological research show that blacks metabolize antidepressants more slowly than Caucasians and Asians. As a result, levels of the medication can build up and make side effects more likely. To be sure, not every African-American is a slow metabolizer of antidepressants; only 40 percent are.

    Notice what’s missing there: the percentage of Caucasians and Asians who are slow metabolizers. If X percent of them are, then Dr. Satel is arguing those patients should suffer greater side effects so that (100 – X) percent can potentially have quicker resolution of their depressive symptoms. If X = 0%, that’s totally cool, if it’s 40% or greater it’s most certainly not, but what if X = 5%? Are we justified in letting 25% of Caucasians/Asians suffer greater side-effects to benefit the remaining 75%? Dr. Satel left out critical information, which is important because she’s the only person I can find who’s argued for different doses due to race.

    And we have reason to think she’s wrong, too. Some digging brought up a 2006 article on clinical guidelines for prescribing anti-depressants, based on allele frequency.[1] Here’s a summary of Table 2, which covers genetic variants found in various races (UM = ultra metabolizers, PM = poor metabolizers):

    Caucasians: 1-10% UM, 5-10% PM type A, 2-4% PM type B
    East Asians: 0-2% UM, 1% PM type A, 10-25% PM type B
    African Americans: 2% UM, 1-2% PM type A, 1-5% PM type B
    North Africans/Middle East: 10-29% UM, 2% PM type A, 2% PM type B

    That article doesn’t give race-specific diagnostic criteria, because there isn’t any good advice to give, the differences between races are small and with huge error bars attached. Instead, it pushes targeted genetic testing that would deliver much better results that assessing someone’s skin or ancestry. Note too that it claims about 4% of African-Americans as poor metabolizers, not 40%, and that if any race could be considered a poor metabolizer, it would be Asians.

    I also have to wonder how good a doctor Satel is. The most common side-effects of Prozac are hives, restlessness, itching, and rashes. The cost of delivering the wrong dosage is pretty trivial. Wouldn’t it have been better to prescribe the same dosage to everyone, then ask people with those symptoms to alert her immediately so she could decrease the dosage? That way, those who respond well to the typical dosage get their faster relief, while those that don’t only suffer for a few weeks.

    I could go on, but I think you get the gist.

    [1] de Leon, Jose, Scott C. Armstrong, and Kelly L. Cozza. “Clinical guidelines for psychiatrists for the use of pharmacogenetic testing for CYP450 2D6 and CYP450 2C19.” Psychosomatics 47.1 (2006): 75-85.

  20. Kevin Kehres says

    @18 Pteryxx

    Whites have better lung function?

    Despite the fact that Kenyans and Ethiopians are demonstrably the best marathoners? And there hasn’t been a white winner of the Olympic 100 meter dash since 1972? And there hasn’t been a white cornerback in the NFL since forever?

    Innate, my lily-white ass…try poverty, pollution, and other asthma-triggers.

Leave a Reply

Your email address will not be published. Required fields are marked *