Yes, but.
Meritocracy in science is already tainted with inequalities
This is similar to the long-standing debate of whether affirmative action is “good” for whatever the goal is for an institution, or if it’s “fair.” The answer isn’t as simple as a “yes” or “no.” It’s a multi-dimensional problem, which is why this essay is longer than a single paragraph. Science has never come close to being a perfect meritocracy. Empirically speaking, talent and hard work are often indistinguishable from nepotism and cronyism. This is true in every industry, perhaps most notoriously in showbiz, and is less common in science, sure. However, this notion of science being perhaps the most meritocratic discipline may have also caused some to believe that science is immune to these inequalities, like an Olympian who thinks he doesn’t need vaccines because he is healthier than everyone else. Moreover, because of all the inequalities that already exist in society, even a “pure” meritocracy in science would reflect the same inequality. For example, the vast majority of science Nobel laureates have been white men. It is simply a reflection of the global societal hierarchy. But what’s wrong with that? Why should science sacrifice its meritocracy to fight this empirical inequity?
Empirical imperfections require empirical interventions
There is a romantic and naïve narrative that science is a pure pursuit of knowledge and therefore does not have to “answer” to society in any other aspect. This may be true hundreds of years ago when science was a vanity past-time for aristocrats to satiate their curiosity — a common narrative that is still taught to school children, e.g. the stories of Newton and Galileo, etc. But, today, science is funded, publicly or privately, to achieve a specific goal that warrants the funding. And while curiosity may still be the main driver for many scientists to get into research, it is not what they are paid to serve. They are paid to serve society, the society that hires them, and society has more needs than just to progress knowledge — even when it is funding science. There are plenty of experiments that can advance knowledge but can’t and shan’t be done because of ethical reasons, e.g. human experiments, etc. So, science should be no stranger to having to abide by societal preferences. If it is in the interest of the society to treat certain hierarchical woes that have also tainted the people of science, it is the duty of those who work in science to fix these problems.
This is not controversial, at least for one side of the coin. For even the most conservative pundit would (probably) not object to firing a scientist not based on merit, say, if the scientists is a rapist. The other side of the coin, however, is more controversial: Is it also acceptable to hire a scientist not based entirely on merit, say, to boost the rating of a candidate who belongs to a disadvantaged group? By the way, some people may argue that we should promote diversity because it is ultimately “good” for science, that the person who will cure cancer may come from poverty, etc. This is a bad argument. It is along the same line that we should pursue green energy because it may ultimately be “good” for the economy. The point is that this is something that needs to be done regardless of the economy, or in this case, regardless of if it hurts the meritocracy of science. The goal is to promote equity and try to be better at this than the status quo set by society — to be part of the collective effort to right this wrong and figure out the cost.
The next few sections are about where I think scientific publishing fits in this conversation. You can skip to the end if you find this boring.
Is publishing too late in the pipeline to contribute?
It is later than most. All of these problems are best treated at birth, when individuals are the most equal, before inequalities have the time to be baked into inequity. It is far more controversial to treat equity than equality, because to treat inequity, you often need to pay with inequality, e.g. via affirmative action. But this is not a reason to not try to treat inequity. Otherwise, we are essentially giving up on everyone who has already been born and affected by inequality. Too bad that you were one of the runners who had to crawl the first 10 miles of the marathon, but we’ve removed this inequality for the next race, better luck next time! I’m sorry but we won’t be helping you by giving you an unfair advantage over the other runners who didn’t have to crawl, because that would be unfair to them. Equality and equity interventions work on different timescales. Equity-oriented policies are bandages for treating where equality policies have failed, for a price in inequality. By the time researchers reach the career stage to publish, much of the inequity has already set in. For example, even if a publisher accepts every physics paper by a female author, it probably still won’t hit a 50/50 target. But even that’s not possible, because a publisher cannot publish bad science. This is easier to explain using some numbers. For example: There is a 100 papers per year quota for the publisher. Each year they get 450 submissions from male authors and 50 from female authors. A 90/10 split. Let’s say 50% of all papers pass peer view, which is generous. Publisher adopts the most aggressive equity policy and publishes every female-authored paper that has passed review. So, 50 x 50% = 25. The rest of the quota is made up by the male-authored papers, so 100–25 = 75. The final acceptance rate for male authors is 75/450 = 1 in 6. The final acceptance rate for female authors is 25/50 = 1 in 2. The system is decisively unfair to male authors. But the final male/female split is still 75/25. This is the price, for how much equity one can buy with inequality. And the more inequity exists in a system, the more costly it is to purchase equity with inequality. Right now, publishers get to choose whether to pursue certain diversity goals. And the lack of transparency when it comes to these decisions gives publishers some discretion about these choices. Because if a publisher decides to make a diversity policy public, it may be challenged in court, which the publisher may not have the resources to fight. (More on the topic of “undemocratically selected curators” for another time, which will include journalists as the largest group.)
Another question is the granularity of these policies. For instance, some research fields may have not enough female white authors, while some may not have enough male black authors. So, how granular should each publication aim for in their diversity policies?
The demographic asymmetry between scientific publishing and science
There are a lot more women in scientific publishing than in science, and there are a lot more non-US/non-UK-based individuals in science than in scientific publishing. Should publishers try to best reflect the demography of science, similar to how Congress should try to reflect the demography of the population it represents? Or should they try best to reflect the demography of the general population? Either way, I don’t think you will find much resistance from science publishers, which are generally rather liberal organizations, to implement diversity goals in their hiring regarding gender or race. But when it comes to nationality, it is different. Unlike gender or ethnicity, immigration status is not a protected class, and there are immense barriers for non-American/non-European nationals to take part in science, e.g. they can’t obtain a working visa even if they are the best person in the world for the job, or they don’t get recognized because they do not work for an institution with name-brand recognition among publishers, which are mostly US- and UK-based, etc. So, one may say that immigrants are not only not a protect class, but are actually explicitly discriminated by law. Just ask any non-American/non-European scientist. My own experience regarding this issue is too long to include here, and it’s probably the same story that you can get from any immigrant. If publishers are passive to these restrictions, they will only enforce these trends. Ultimately, this is part of the larger problem of cultural hegemony, which has a self-reinforced mechanism, and remains as a blind spot for most intellectuals in the US when talking about inequality.
Epilogue: Where are we now?
It will be nice to say quantitatively where we are and where we should be — in terms of if we are currently “investing” too little or too much when it comes to exchanging equity with inequality. However, I don’t think we need to be quantitative. We only need to say qualitatively, if we are trending in the right direction, and I’d like to think that we are (when it comes to bring equity to the “protected” classes of people, e.g. gender and ethnicity), but we also are not (for those whose disadvantage come from an “unprotected” class, e.g. nationality and socio-economic status, etc.)
Societal changes may also be increasing inequalities faster than current policies can keep up. Because even with all the gender- and ethnic-centric policies in place, their impacts can easily be dwarfed by the fast-growing economic inequality, so these policies may have no significant impact macroscopically, no more than an aesthetic patch that benefited proportionally few individuals. But the notion of explicitly tackling economic inequality may be too unpalatable in US politics, and so we are left with these band-aids. That’s the pessimism talking, but I think that’s closer to reality. Liberal optimism can be blinding sometimes, and it is often better than nothing, and “better than nothing” is often the bar for these things. This time, it’s the cynicism talking.
