Monday, May 13, 2013

Leigh Alexander and I agreed to move the chat we started via Twitter to email so that we could express and explain ourselves without the character limit.

Leigh Alexander and I agreed to move the chat we started via Twitter to email so that we could express and explain ourselves without the character limit.  As Leigh tweeted, "Disagreement or the need for further discussion can't preclude respect among colleagues." The conversation:

Leigh Alexander: So I saw you as being displeased that people too easily leaped to attribute the wage gap to sexism, because they were eager to earn accolades for being politically-correct, and that they saw sexism where there might not be any in order to feel they'd won a moral victory.

Shawn Elliott: What bothers me is that they both attribute the wage gap to sexism on the part of game industry employers by arbitrarily banishing any serious consideration of alternative explanations from the realm of possibility, and proceed to issue moral imperatives, demanding that the accused address wrongs for which they haven't actually been convicted. In other words, when someone admits that they don't understand the actual conditions that contribute to an unfavorable outcome, then sentences someone all the same, and explicitly tells the accused that their “only valid response” is to rectify the perceived problem, they're moralizing plain and simple. I don't doubt or call into question their commitment to equality (more on this in a moment). I believe their commitment is sincere, and ask that you accept that mine is as well. But it's possible to come to dumb conclusions despite having your heart in the right place.

LA: On one hand, I completely agree with you that if we're going to address representational problems financially and otherwise, we do need a complete picture of the disparity, in order to make a stronger case that it can't be explained by other means. Logical fallacies don't help anyone's case.

SE: We need a more complete understanding of the causes of the disparity in order to correct if it is indeed due to discriminatory policies and practices. You can't fix what you fail to comprehend, no matter how nice it makes a person feel to pretend that they have sufficient info to tell a particular party that its their fault and that they need to fix it. If it turns out that the problem is more challenging and multidimensional, the finger pointing will not have helped anyone at all – in fact, it can contribute to the problem. It can lead you to overlook much more promising avenues for change. Poor reasoning can allow people to dismiss your concerns as well as your diagnosis. I'm taking pains here to clarify that I do not dismiss the underlying concern.

LA: On the other hand, that women are unequal in the games industry is a fact of life for us. Sexism, hostility and general alienation are the reality for many if not most of us. That isn't something that requires data or is up for debate -- women in the industry have been trying to talk about our experiences at increasing volume and at critical mass over the past few years. Many men will find any reason not to listen to us, but we are trying.

SE: I absolutely understand this, however, I addressed wage disparities among developers alone, not the treatment of women as a whole within development, and definitely not across the industry spectrum of reporting, blogging, streaming, forum-going, and gaming.

LA: So I don't think it's actually such an intolerable leap for some people to have seen that data and presumed some degree of logical correlation. Women are unequal in the games industry; women's pay is also unequal. You are right that we don't know the extent to which one fact causes the other, and indeed, it would help to know. 

SE: It isn't a tolerable leap to say that, since women are exposed to persistent sexual harassment within games, for example, wage disparities among developers are necessarily due to sexist practices and policies on the part of employers. When you write that “Women are unequal in the games industry” do you specifically mean that devs are disproportionately male? (I'll pass on the pay aspect, as that is the point of this discussion.)

LA: But the correlation is a fair assumption. I would trade my right hand if further research into the pay gap found that sexism does not play some role (especially as you and others discussed women being relatively new to the industry or less-experienced -- why would that be the case, for example?). It is absolutely, at least, an abstractly-related discussion even if the data doesn't "prove" it.

SE: Again, my argument is not that sexism is unlikely to play a part in explaining why fewer women than men work in development, or even why on balance women developers are paid less than their male counterparts. When I wrote that “We must also study the male-female ratio among qualified candidates from whom studios receive applications. We must ask what percentage of women pursue educations in programming?” I meant to imply that sexism may enter the equation at any number of steps on the path from a woman's childhood experiences and environment up to her ultimate decision to pursue a career in development. I also cited the well-documented effects that marriage and career interruptions have on occupational compensation (and how this differs depending upon one's field of expertise), and sexism may play some part here, as well. But that blame is not necessarily the employer's. And that doesn't scratch the surface of the meaningful, multidimensional discussion occurring outside the walled garden of gaming industry coverage. (See Po Bronson and Ashley Merryman on gendered competitiveness or Jennifer Lawless and Richard Fox on political ambition )

LA: And people are righteous about it because a certain degree of outrage has accumulated in the discussion on sexism, and justifiably. Whenever we talk about sexism, a man appears to find some way to either abuse or discredit us (even when "we" includes our men friends, who also increase their volume of outrage). A common derailing tactic when someone has a hostile political agenda is to play the "I'm simply discussing the facts" card, to veil the fact they're refusing to have empathy for the party that's requesting it.

SE: I find it a bit embarrassing -- as though I'm being put on trial – to need to explain that I'm wholeheartedly for equality and intolerant of sexism. Every woman who faces sexism (every woman alive?) has my empathy. At times, though, you do need to discuss known facts (and challenge unsubstantiated supposition), and doing so has nothing to do with evincing empathy.

LA: The numbers in the salary survey are not the entire conversation, you are right. But they're the beginning of a conversation, and I would have liked to see you frame your argument as asking it to go further, rather than discrediting others in your desire to prize facts or further research. At the very least, it shows more care toward underrepresented people if you presume their instincts about discrimination may be correct and you seek data to support that, rather than asking for data before you will consider their feelings.

SE: I accept blame for my poor writing. It was my intent to expand the scope of what I saw as constrained conversation, not to close it. With available data, though, I can't presume that instincts that say employer discrimination explains the wage disparity in game development are correct. They might be, they might not be. Not coming to conclusions about causes before adequate evidence is available doesn't preclude the ability to consider someone's feelings. I don't understand this argument at all.

LA: Because the statements made that you criticized -- things like how we need to train, hire and promote more women -- are pretty important goals for the industry that I don't think you'd disagree with. I was disappointed in your eagerness to banish them from a conversation about wage gap when even in absence of statistics, we know there is very probably some relationship.

SE: Did I criticize the notion that “we need to train, hire and promote more women”? I remember writing that “These are fine policies to follow.” That said, more women working in game development is a great thing – same for members of other groups that are less represented than women by far.

LA: Border House, or RPS, are allies of diversity. You say you want to enable them to be better allies through more information, but your clear disdain for their logic can unfortunately be read as derisive of their long campaigns for equality.It's a destructive rather than supportive approach.

SE: I'm not aware of their campaigns and was responding to what I read. I didn't intend to deride the spirit of their campaigns. And you're right in that my tone was too combative. I think the tone I took from the stories was “I'm going to tell you how it is and don't you dare argue or I'll dismiss you for who you are and not what you say.” I took that as a challenge and it immediately turned me off to RPS's approach in particular.

LA: Maybe I'm misreading you some, but then I'm not the only one. As it is now, you have commenters whose takeaway may be that allies of diversity are just getting uppity to score political points, or that someone like John Walker is insincere (men who champion against sexism are often accused of 'white knighting' or pretending to care in order to get laid or something, so your insinuation that these kinds of articles try to 'score points' is weighted). Or that "we need less outrage," etc.

SE: Fair enough. I hope that this discussion – if we publish it – will help clarify my perspective. The logic was so loopy that I supposed politicking or sheer naivete was at play. I found the former assumption more charitable. Again, my tone was too combative. And as I wrote on Twitter, “Suggesting that the potential causes for an outcome are more complicated than a specific speaker insists they are in no way says that emotional response to that outcome is improper. I'm not saying we shouldn't be upset by it so much as I mean to say that we shouldn't prescribe solutions when when we aren't certain as to the causes.” Although to add to that, I disagree with directing a strong emotional response at a group (“those in senior positions at publishers and developers”) before their guilt's been decided.

LA: When I worried you'd arm our enemies, what I meant is that every time a woman writes an essay about her personal experiences, there are commenters complaining she is emotional, irrelevant, lacks 'facts' to 'prove' she has been a victim of a situation, and so on. The commenters say things like 'we need to stop all this emotional manipulation and stick to games' or whatever, as a way of silencing people. And now they'll be able to grab your essay, which is about facts over outrage, and which can be read as attributing vocal writers' passion to an insincere 'political agenda', and use it as ammunition in that kind of argument, which I'm sure is not how you wanted to come across.

SE: I agree, anyone who dismisses concerns about sexism because a few writers used poor reasoning in one place is dumb. I should underscore that. I imagine you've encountered plenty of people who were willing to dismiss feminism as a school of thought because Harding characterized Newton's Principia Mathematica as a “rape manual” or because Irigaray argued that E=MC2 is a “sexed equation” or that “masculine physics” neglects fluids. (Not that the arguments I addressed approach that now legendary lunacy. ) But we can't be too timid to critique an argument simply because the sentiments of those who make them are in the right place. Another example: I believe that sufficient empirical evidence exists for me to be concerned about anthropogenic climate change, and for that reason I was upset by “Climategate” before it came to light that the controversy was manufactured.

LA: You totally could have made your point as regards all the things we have yet to know about the salary data, and why it's important for us to know them (so we can build better, more positive arguments for this equality we all want). in a way that did not target and dissect others who were only trying to express that commitment, albeit imperfectly.

Does that make sense? I think if you're joining the gender inequity conversation you do want to be very sensitive about the format you choose, and the tone and purpose of your argument, and which sides in the discussion you are really best enabling and expressing support for. Making fun of RPS for assuming 'moral DEFCON' just equips people to think sensitivity is something suspicious or worthy of ridicule. I really don't want to believe that was the tone you were going for.

SE: It isn't. See above.

LA: This isn't, unfortunately, the kind of thing we can treat with the same dispassion with which we handle other journalistic issues. We want to teach others to model compassion, too, and mobilize them to care about equality on every frontier, not to feel justified in defaulting to skepticism in every conversation. They already do that enough.

So many people look up to and listen to you. Please teach your readers that being angry about sexism, and wanting to enable more women to participate in the industry, are meaningful, honest and moral positions to take, not the province of the self-serving or uninformed. And that the reason we need more data is so that we can fight these battles better, not so that we can debunk outrage and passion.

SE: I understand and appreciate your concern. 

Thursday, May 9, 2013

A topic that deserves to be talked about, and for that reason we should subject morally attractive propositions to more scrutiny, not less.

Moral posturing is so alluring that it leads us to neglect normal standards of logic. Take, for instance, RPS assuming the moral DEFCON 1 alert posture.
I hope the use of “horrendous” and “despicable” (the source story, still at a slightly lower DEFCON, settles for “disgusting”) stand out, as well as the dogmatic assumption that statistical disparities alone demonstrate discrimination. My point is not that sexism may or may not play a part in the wage disparities that Game Developer Magazine's data allegedly documents, and it definitely isn't that the industry shouldn't pay matching salaries to male and female employees with the same experience and skill sets. Rather, I'm writing to challenge the uncritical assumptions on which both stories are built, and way that the opportunity to publicly sit on the side of the angels seems to contribute to the error. Because both stories note that Game Developer Magazine's numbers offer no information on the years of experience specific to the men and women surveyed, but then arbitrarily banish any serious consideration of alternative explanations from the realm of possibility in order to prescribe the usual preferential policies.

There will be factors, certainly,” RPS writes. “The hugely larger numbers of men in the industry means by nature there will be many more of them who have worked for longer, and thus secured ultimately higher salaries. But there are women who have been involved for a long time too, and this absolutely doesn’t explain away these massive discrepancies.

I trust that the sheer absurdity of the argument speaks for itself.

Similarly, BorderHouseBlog says, “I’m sure there are more details that might make these numbers less damning.  For example, we all know that games have been long dominated by men and the industry is taking small steps to change that.  As a result, many of the women who answered the survey might be new to the game industry, might not be in as senior of roles as the men who responded.” And then, in a comically clumsy attempt at sleight of hand, BorderHouseBlog continues: “However, I don’t think this changes the fact that we need to recruit and encourage more women at all levels of every organization — and we’re failing to do so.”

In other words, the writer acknowledges that while the evidence she predicated her argument on cannot support her claim, we are to accept that that now discarded claim is relevant to another new and equally unsubstantiated claim, no burden of proof required.

And next come the aforementioned prescriptions for preferential policies. As RPS has it, “the only valid response is for those in senior positions at publishers and developers to not pretend it isn’t them, to look at their own figures, and to rectify discrepancies.

This is ridiculous. Since when do we draw conclusions without a scrap of empirical evidence, admit that we lack an argument to stand on, and then issue moral imperatives, demanding that the accused address wrongs for which they haven't actually been convicted? We haven't even determined with certainty whether or not there are discrepancies, let alone ascertained causes should their existence be confirmed. Anyone attempting to explain why more men than women work in a particular industry (indeed a question well worth the investigative effort) has laborious analytical work ahead of them. But it's easier for ideologues to attack potential bogeymen and behave as though they've done their part to change undesirable circumstances than bother to consult the work of those who do such as Cornell professor of economics Francine D. Blau.

Again, to my point, it's all a muddle of poor thinking in the race to righteous outrage. If you were among “those in senior positions at publishers and developers” and believed you could pay group A some fraction of the wages that you pay group B and obtain precisely the same results, plus stand to benefit politically from appearances in the bargain, why would your hiring practices not understandably favor group A at all times, especially in the current economic environment?

Similarly, all of BorderHouseBlog's prescriptions assume that any statistical discrepancies, should they be discovered, are prima facie the result of sexist policies and, as such, can be eliminated entirely by employers: “Leadership: look at your organization. Compare the salaries of the women to the men who work at your company, and align their salaries. If all of your women are junior, evaluate them. How long have they been junior? Are they deserving of an increase in role, capabilities, and salary?”

These are fine policies to follow – notwithstanding the seeming suggestion that employers arbitrarily align salaries according to gender status irrespective of employment history, skill set, hours worked, etc. (In other words, even if an employee has three years of experience and is due for advancement, this will do nothing to equalize salaries if a significant proportion of her male coworkers have fifteen or more years experience a piece.) On their own, however, these fine questions aren't enough. We must also study the male-female ratio among qualified candidates from whom studios receive applications. We must ask what percentage of women pursue educations in programming? What percentage of women whose salaries we survey are married? (Academic women who never married average slightly higher incomes than male counterparts.) How many of the women whose salaries we survey take extended time off for maternity leave? (Different occupations have different rates of obsolescence for their respective skills, so that interruptions of careers in some fields are more damaging to one's career than in other fields.) And the answers that these inquiries yield prompt more pressing questions.

This is a topic that deserves to be talked about, and for that reason we should subject morally attractive propositions to more scrutiny, not less.