I’m totally convinced that ∃ more connections between mathematics and the humanities than the university culture I once stewed in would suggest.

Probably due to personality differences, but also lack of familiarity with each other’s subject matter, I never saw inter-departmental collaborations and—as I’ll discuss in another post—even the idea of data is seen as a four-letter word in the gender studies department. (Likewise, ethnography and anecdote are four-letter words within the economics field, and statisticians also concern themselves only with structured data.)

Nevertheless I see mathematical shapes all over cultural analysis, and I mean to record them. (However typing up a coherent few paragraphs, let alone adding drawings, takes several orders of magnitude more time than simply thinking a thought.)

 

After reading her essay crowing that millennials do not see themselves as special, I went on to read more of Phoenix and the Olive Branch, which talks about rehabilitation from “Quiverfull” fundamentalist upbringing—particularly gender issues that arose as a Quiverfull young woman.

(Relevant to the “value of liberal arts" question, Sierra writes that "College literally saved my life"—without the critical thinking skills—not science or programming skills—that she learned at college, her mind and heart and … uterus would have remained ensnared in the “Quiverfull” fundamentalist mindset she grew up in. Just an interesting sidelight.)

 

Sierra has a very logical way of describing a flaw with sexist views:

Check out this gem from “Reclaiming the Mind”:

You see, when people are truly committed and consistent egalitarians, they have to defend their denial of essential differences. In doing so, they will advocate a education system in the home, church, and society which neutralizes any assumption of differences between the sexes. In doing so, men will not be trained to be “men” since there is really no such thing. Women will not be encouraged to be “women” since there is no such thing. The assumption of differences becomes a way to oppress society and marginalize, in their estimation, one sex for the benefit of the other. Once we neutralize these differences, we will have neutered society and the family due to a denial of God’s design in favor of some misguided attempt to promote a form of equality that is neither possible nor beneficial to either sex.

As a truly committed and consistent egalitarian, yes, yes I do deny “essential” differences. You know why? My essential nature is not “woman.” My essential nature is me. Sierra. It’s who I am. …[M]y best friend[’s] essential nature [is] not identical to mine. It might have similar colors and shapes, but so would mine and my fiance’s. Because people are different. “Men” are not more different from women than they are from other men.

In statistical or mathematical language, I would interpret this as saying "The fact that gender==Woman is not entirely determinate of everything about me.”

If I were writing a computer program to mimic the kind of sexism Sierra is talking about, it would take one input for gender and, if the answer is male, then prompt for further details on the personality, achievements, background, interests, thoughts. Elsif gender == female, then the only questions worth asking are “Fat? Hot?" Otherwise, break; because there is no else.

Not that the "Being a minority is determinate of everything and only males can show variation” is limited to gender. On Reddit we find:

"I can’t imagine a black guy saying ‘anywho’"

as if blackness is somehow so determinate of behaviour. Charmed, I’m sure.

 

In statistics the paradigm is that data go into a model and a couple numbers come out. Some of the numbers parameterise the model. But other numbers tell us how good the explanation is. There are numbers to tell us how well individual parts fit, how well the overall whole fits, and several numbers that are warning indicators for various types of traps that can make the other numbers mess up.

Thinking that everything about a minority is determined by their minority status is a bit like ignoring all the model-fit numbers.

If we explored some data with a large number of linear models, progressing from coarse (few terms) to fine (many terms), we would probably see gender differences as a significant term among coarse models. But those models would also have a low specificity and explanatory power. Then as we added more explanatory terms (finer models), those other explanators—correlates of gender/race, but not gender/race itself—would start to steal explanatory power away from the gender  dummy variable.

To give a physical example, 100m sprint times show differences across male/female, but training is more determinate of the sprint time. If we could measure personality and thoughts and the kinds of traits that Sierra might say define her as a person, we would probably be left with very little t-value on the gender dummy.

 

One more mathematical parallel. The idea that “minorities show no variation; only the privileged group can be variable” is isomorphic to Jim Townsend’s mathematical-psychology model of racism. Substitute “minority” with “other group” and “privileged group” with “self" or "my group" and you have the same model of a negatively curved metric space:

negatively curved metric space for self versus other (privileged group versus minority group)

So you thought postmodernism was opposite to science? Here is Derrida’s “privileged hierarchy” where “one term dominates the other” — at least one mathematical interpretation of those words.

91 notes

  1. clazzjassicalrockhop reblogged this from isomorphismes
  2. bakomglaset reblogged this from isomorphismes
  3. saschaisking reblogged this from proofmathisbeautiful
  4. immusical reblogged this from proofmathisbeautiful
  5. bridgeacrosstheuniverse reblogged this from isomorphismes
  6. tmbird reblogged this from proofmathisbeautiful
  7. uncatchablefox reblogged this from proofmathisbeautiful
  8. idlnmclean reblogged this from isomorphismes
  9. teannoms reblogged this from proofmathisbeautiful
  10. budgettravels reblogged this from proofmathisbeautiful
  11. jpallhvarast reblogged this from proofmathisbeautiful
  12. geometricespo reblogged this from proofmathisbeautiful
  13. tiradia reblogged this from isomorphismes and added:
    It’s so refreshing to read people applying logic and mathematics to refute simplistic assumptions.
  14. mainemunky reblogged this from proofmathisbeautiful
  15. withagrainasalt reblogged this from proofmathisbeautiful
  16. notafraidtosayicare reblogged this from proofmathisbeautiful