Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!apple!usc!orion.cf.uci.edu!uci-ics!tittle
From: holstege@polya.stanford.EDU (Mary Holstege)
Newsgroups: soc.feminism
Subject: Re: sex/gender
Message-ID: <10546@polya.Stanford.EDU>
Date: 13 Jul 89 00:23:33 GMT
References: <8907071844.AA10158@cattell.psych.upenn.edu>
Sender: news@paris.ics.uci.edu
Reply-To: Mary Holstege <holstege@polya.stanford.edu.stanford.EDU>
Organization: Stanford University
Lines: 210
Approved: tittle@ics.uci.edu

In article <8907071844.AA10158@cattell.psych.upenn.edu> you write:
[Gretchen wrote, that is --clt]
>
>	I would like to argue that there are indeed differences
>between men and women, but that these differences are not necessarily
>(morally) valid or unchangeable.
...
>gender maps on to evironmentally determinined behavior.  The first
>point I would like to make is that characteristics about me which are
>the result of my genes are no more a part of me, no more natural, than
>characteristics which are the result of my experience (learning,
>nutrition, etc.)  A second point is that there is no such thing as a

Of course, this is undeniably true, but to ignore the political force
of the term "natural" as applied to alleged sex-differences is, to put
it politely, dangerously naive.  When someone like EO Wilson says that
women are "naturally" less agressive than men, he means that women are
genetically doomed to lose in competition with men (in jobs, in
money-earning, etc.).  He says that a code of ethics based on such
supposed facts "would be genetically accurate and hence completely
fair."

Claims that women and men differ "naturally" in one way or the other
are primarily used to argue against any attempt to improve the lot of
women, or to argue for limiting women's choices.  Advanced education
would ruin women's reproductive systems; hard work makes them
infertile; women can't do math so no point wasting resources teaching
it to them; women are betting are small motor skills so they should
take up typing and knitting; etc. etc.  The litany is long,
depressing, and hasn't changed much.

>	There certainly are statistical differences between men and
>women.  More women bear children than men, but also, women are more
>verbal and less spatially skilled than men; women are more nuturing
>than men, less sexually aggressive than men, etc.  

There are two general comments I want to make about this list of
`facts'.  First, the `facts' as stated are not what has been
statistically shown.  Let's take the `facts' one by one.  The first is
undeniable and a fact of definition -- women are those members of the
species who have certain reproductive organs.  As for the rest...

First note how the `fact' is stated "women are more verbal than men."
The natural reading of this is that most women are more verbal than
most men, with the heavy implication that each woman is more verbal
than each man.  Actually, all that has been shown is that, in those
(minority) of studies that show a difference in the performance of
(overwhelmingly college-age, white, American) men and women, the
average of all the women's scores is slightly greater than the average
of all the men's scores.  Most studies show no difference at all, and
meta-analysis of the data shows that only 1 percent of the variability
can be accounted for by sex.  That is to say, 99% of the difference
between two people in this test has nothing to do with sex.  Further,
this information tells us nothing about how "natural" this difference
is, although such evidence is commonly cited for just such purposes.

Suppose we have a test of vocabulary use.  Suppose one of the items is
`barrette', which refers to something that in our culture women tend
to have more familiarity with than men.  Suppose we give this test to
a bunch of people.  Some of the males will get the item wrong, some of
the females will, but on average more males than females will get this
item wrong.  Let us suppose that the remaining items are unbiased.
Then we will have `proved' that women are naturally more verbal than
men.  The alert reader will notice that we could prove just the
reverse by putting in a lot of vocabulary terms that relate to
baseball and automechanics.  It is also interesting to notice that
this alleged difference has disappeared in the SAT realm, because ETS
has decided that if men do worse on a test there must be something
wrong with the test.  More on this in a moment.

Next we get to the alleged superiority of males when it comes to
`spatial reasoning'.  One of the favoured tasks for testing this
ability involves putting the subject in a dark room with a male
experimenter and asking the subject to judge when a light bar that is
skewed with respect to a lighted frame is vertical or horizontal.  The
experiment is often run by having the subject ask the experimenter to
adjust the bar one way or another.  Is this really a test of spatial
reasoning, or mightn't some sexual-social dynamics have a lot to do
with this?  Also note that when spatial reasoning tests are given to
Eskimos who live in an environment with few obvious location cues and
whose culture encourages girls to run around and explore, guess what?
No difference.  The difference generally appears in cultures that
restrict girls' exploration of their surroundings, keeping them close
to home.  Also note that in drafting classes the males start out
better but the females quickly catch up with training.  Could it be
that playing with blocks and climbing trees teaches one more spatial
skills than combing Barbie's hair?  And yet, this difference (5% of
the variability this time) can be and is used to argue that women
(generally, universally even) cannot be architects or engineers, that
it is a waste of time trying to teach them these things.

Related to spatial reasoning, for some reason that escapes me, is
mathematical ability.  Here the so-called evidence is even flimsier.
Most testing is done using the PSAT, the SAT, the GRE, or some other
product of ETS.  These tests are designed and supposedly validated to
predict academic performance.  Yet it is known (and has been for some
time) that these tests consistently underpredict female performance.
So the test is flawed for its stated purpose (which has serious
implications for the academic opportunities of women) and ETS has
failed to fix the flaw, although the difference in performance in the
verbal tests *have* been removed (by conscious choice, by the way).
Why?  Yet time after time researchers give the tests to people, note
there is a (small) sex-difference and conclude that men "naturally"
have more mathematical ability than women.  Again, why?  It doesn't
take a genius to figure out that there is a nasty anti-feminist
political edge to all of this.

Finally, we get to the sexual aggression `fact'.  I know of no
credible evidence to support this.  One can also consider that in
earlier ages and in many cultures today "everyone knows" that women
are *more* sexually aggressive than men.  It should come as no
surprise that this `fact' was the justification for keeping women out
of the public domain, for chastity belts, for harems, for the veil.

Some of these
>differences are larger than others, and some are better backed by data
>than others.  But if someone provides empirically evidence about a sex
>difference, I refuse to close my eyes and pretend it isn't real.  I
>also do not believe that some differences are more "natural" than
>others.  If a difference is found, it is interesting to explore the
>reasons behind this difference, but I don't see why some reasons are
>more natural than others.

This is all true to an extent, but it is important to understand that
many, many, many sex differences have been `discovered' throughout
history, and the evidence to back it up has been feeble at best, and
the differences slight.  It is also true that the differences are
claimed as "natural" hence "immutable" (a non-sequiter, but a popular
one) and that the difference has always been rationalized to favour
certain social agendas.  It has also been the case that `facts' put
forward have flipped-flopped depending on whether current theory made
a fact support male superiority or not: men have less-lateralized
brains than women/less lateralized; men have larger corpus calloseums
than women/ smaller ones; men have larger temporal lobes than
women/smaller...

Even a cursory study of the history of this field should give anyone
pause before swallowing the latest "science proves innane
sex-difference" line.  I don't think one should close one's eyes to
facts either, but one should be very suspicious of those `facts' and
even more suspicious of jumping from a small statistical difference in
some test to a generalized claim that "men are better/worse at X than
women." (implication: naturally, innately, immutably)

First, claims of natural differences can lead to reinforcement of
those claims.  I know many a young girl who, experiencing some
short-term difficulty in math, says "well, girls just can't do math"
and gives up (or would if allowed to).  What do you suppose happens in
a classroom when a teacher believes that women are naturally inferior
at math?  It is naive to suppose that that teacher will not invest
more time in the supposedly more able students -- the males.  Such
effects have been documented.

Second, casting the differences as being between men and women leads
people to suppose that those differences are innate rather than
reflecting the generally different experience of males and females in
our culture.  Such a supposition leads to the idea that there is
nothing to be done, that women will always be paid less, or fail to
advance to certain levels of management, or fail to penetrate certain
job categories.  It doesn't *really* follow, of course, but look at
what people say, even the people who do the studies and presumably
know better.  (Some examples follow.)

Finally, the focus on what are, even taken at face value, very minor
differences leads to damaging results, and a failure to see what the
important variables in differential performance are.  Such a failure
can have important public policy consequences.  What is irksome about
so many of these studies is that they do not control for environmental
effects that are *known* to have a far greater impact on performance
differences than sex.  Consider a related example: there is an
important difference between the statement "blacks do less well on IQ
tests than whites" and the statement that people who suffered from
malnourishment during development and whose mothers lacked proper
prenatal care do less well on IQ tests than others.  If one merely
reports the first result, without controlling for the factors given in
the second, one has learned nothing, because in this society more
blacks live in conditions of poverty than whites.  Worse, the public
policy of the first statement is to shrug one's shoulders and say
"gosh, ain't it a shame those black people are inferior."

In conclusion, a quotation to highlight the importance of the misuse
of data and the term "natural":

"Females, on the average, surpass males in verbal fluency, correct
language usage, spelling,... Males, on the average, are superior to
females in verbal comprehension and reasoning, mathematical reasoning,
spatial perception,...  These differences foreshadow the different
occupational goals of men and women."
Josef Garai and Amram Scheinfeld 1986

[Notice that while females "surpass males" in various tasks, males
"are superior".  Notice also that minute differences that account for
only 1-5% of the variability between individuals is enough to
determine occupational goals of an entire sex.]


A couple of books worth looking at:
  Myths of Gender: Biological Theories Anout Women and Men
  Anne Fausto-Sterling, Basic Books, 1985

  Not in Our Genes: Biology, Ideology, and Human Nature
  R.C. Lewontin, Steven Rose, and Leon J. Kamin
  Pantheon Books, 1984

                              -- Mary
                                 Holstege@polya.stanford.edu

ARPA:                            holstege%polya@score.stanford.edu
BITNET:                          holstege%polya@STANFORD.BITNET
UUCP: {arpa gateways, decwrl, sun, hplabs, rutgers}!polya.stanford.edu!holstege