Ask a Female Engineer: Thoughts on the Google Memo

Many of the more reasonable criticisms of the memo say that it wasn't written well enough; it could've been more considerate, it should have used better language, or better presentation. In this particular link, Scott Alexander is used as an example of better writing, and he certainly is one of the best and most persuasive modern writers I've found. However, I can not imagine ever matching his talent and output, even if I practiced for years to try and catch up.

I do not think that anyone's ability to write should disbar them from discussion. We can not expect perfection from others. Instead we should try to understand them as human beings, and interpret them with generosity and kindness.

> I do not think that anyone's ability to write should disbar them from discussion. We can not expect perfection from others. Instead we should try to understand them as human beings, and interpret them with generosity and kindness.

I'm a huge proponent of the principle of charity, but I found it impossible to apply to the Google Memo. Not because I'm deeply mired in political correctness (I have a range of views people in my circle consider right-wing) but because it's so badly reasoned it makes it hard to presume good faith on the part of the writer.

Damore points to studies showing that, e.g. women are more agreeable and more people-oriented. From that, he concludes women on average are less likely to prefer programming. We can diagram this reasoning as follows (the arrow with the line through denotes a contraindicator):

Women -> (agreeable + people-oriented) -> [???] -\-> programming

As you can see, there is an unstated premise:

(agreeable + people-oriented) -\-> programming

Damore's argument thus reduces to a bit of begging the question. We assume that programming is a "masculine" profession. Thus, being agreeable and people-oriented, which are feminine traits, must be contraindicators for preferring a career as a programmer. We have no studies that show this--we just assume it.

Edith, by the way, demolishes that assumption: "For example, students and professors I met in college that grew up in the USSR thought engineering was stereotypically women’s work." That demonstrates how the "gender" of various professions is a social construct. In India, where men are over-represented in teaching, it's not considered a job for "agreeable" "people-oriented" women. It's men's work. Law was historically considered men's work (it's analytical and adversarial, and could be called "people oriented" only if you hate people). But that view has been redefined as more women enter the profession. Likewise for medicine, accounting, etc. Accounting is an archetypally "masculine" profession (locked away in a back closet crunching numbers), but today more than half of accountants are women.

The moral of the story is that if you're going to make a controversial point, it had better be a good point. Damore's memo wasn't just badly written, it was badly reasoned, and deserved the scorn heaped on it.

> Damore points to studies showing that, e.g. women are more agreeable and more people-oriented. From that, he concludes women on average are less likely to prefer programming.

That's Diekman 2010:

https://www.ncbi.nlm.nih.gov/pubmed/20631322

"Although women have nearly attained equality with men in several formerly male-dominated fields, they remain underrepresented in the fields of science, technology, engineering, and mathematics (STEM). We argue that one important reason for this discrepancy is that STEM careers are perceived as less likely than careers in other fields to fulfill communal goals (e.g., working with or helping other people). Such perceptions might disproportionately affect women's career decisions, because women tend to endorse communal goals more than men. As predicted, we found that STEM careers, relative to other careers, were perceived to impede communal goals. Moreover, communal-goal endorsement negatively predicted interest in STEM careers, even when controlling for past experience and self-efficacy in science and mathematics."

That study doesn't support either of Damore's premises. It doesn't address biological tendencies--it was conducted on adult women. Nor does it address traits actually linked to professions. It focuses on perception. That's an important distinction. In the age of the internet, software is much more about fulfilling communal goals--e.g. helping people communicate with family and friends--than many women-dominated professions like accounting. To the extent women perceive the opposite to be true, a strong argument can be made that it is the result of a male-dominated profession characterizing itself as such, rather than anything inherent about the profession.

Peoples' perceptions of various professions are the result of socialization. For example, my mom grew up in a society where teaching was a male profession--it was characterized as being about instilling wisdom and discipline in children. She found it very upsetting that teachers in the US were overwhelmingly women.