I’m acquainted with Holden Thorp, editor-in-chief of Science, because my colleague Luana Maroja debated him and then wrote about it afterwards (you can see the debate video here and Luana’s post-debate interview with a National Association of Scholars person here). Thorp, like every other big-time journal editor, is woke. You can see that in the magazine, but also in the debate with Luana. Writing about the debate at the Heterodox STEM site, Luana said this:

What Thorp does not seem to realize is how offensive it is when it is argued that inclusiveness requires special accommodations, such as lowering the expectations for people who “look like me.”  I described my experience of participating in a training session for a hiring committee at the college where I now teach.  During this session, we were told that “we cannot expect as much from Latina women [as from white men], because they have more obligations towards family,” something I found incredibly insulting, as if I don’t have the agency to decide how to balance my own time just like anyone else.  Other initiatives in the name of inclusiveness, such as chasing microaggressions, are even more negative and damaging to the individuals who internalize this concept – imagine that you adopt the microaggression mindset and live your life thinking that the world is turned against you. Consider this not-that-hypothetical scenario: you walk and wave to a student and the student does not wave back to you.  You have two choices: you might decide that this was a personal microaggression due to who you are, or alternatively, you might conclude that the student simply did not see you.  Only one of these two views can lead to a good life and mental health.

In many instances during the conversation and in his writings, it is clear that Thorp subscribes to a Woke worldview.  He believes in the value of diversity, but assumes that the diversity can be attained only by lowering the bar for women and minorities, and that “inclusion” can be achieved by excluding white males.  Ironically, at least twice during the conversation his comments revealed that he does not consistently apply this logic. Prior to the conversation, when we all showed up on time, he commented that we did so “because we are all scientists” – this ignores the fact that my culture (Latina Brazilian) does not respect punctuality, and that I had to learn to do so for my own benefit.  He then pointed out that, “as scientists we were the first to run with our complete AP calculus tests to our teachers in high school.” Well, in Brazil I had a third-world education… I did not have the opportunity to take calculus until I was in a PhD program at Cornell.  I certainly did not study AP calculus in school, and if I had dared run waving a completed exam to the teacher, I would be sure to never to have friends again… It is a pity that the topic of culture was not discussed more in this conversation – I imagine Thorp’s view would be that “all cultures are equal in their outcomes”, when they clearly are not.

In the end, I was unsure if Thorp is a true believer in the need to lower standards in the name of inclusion, or if he plays a game, where he is a white savior.  It is hard for me to understand why some people, with all good intentions, fail to see the obviously damaging effects of their ideologies and actions.  Lowering standards and expectations hurts the most vulnerable of us; it does not help science or the people that such actions are intended to help — and I hope we can start pushing back hard against this damaging ideology.

Well, Thorp is still riding this horse, as we can see in his “editor’s blog” that went up at Science yesterday. There he created a special post to go after our paper “In defense of merit in science“, as well as after Pamela Paul’s NYT column describing the paper. In fact, these are the only two links he gives in his piece. Click to read:

His point, which could be expressed much more succinctly, is the claim that a more diverse group of people can do better science than a less diverse group. In fact, it’s more than that: he argues implicitly that a more diverse group of people can do better science without having to lower the bar for judging science or scientists.

It’s clear that by “diversity” Thorp means “racial diversity”—as that’s the one example he gives—but he may mean gender diversity as well. He gives a nod to “viewpoint diversity,” but it’s clear that he doesn’t mean, “let’s get more conservatives and poor ‘first generation’ students into science.”

But first, for reasons best known to Thorp, he makes The Argument from Humanity”: he thinks that people like the 29 of us who wrote the merit paper don’t recognize that scientists are human beings, and that this somehow blinds us to the virtues of diversity:

It has somehow become a controversial idea to acknowledge that scientists are actual people. For some, the notion that scientists are subject to human error and frailty weakens science in the public eye. But scientists shouldn’t be afraid to acknowledge their humanity. Individual scientists are always going to make a mistake eventually, and the objective truth that they claim to be espousing is always going to be revised. When this happens, the public understandably loses trust. The solution to this problem is doing the hard work of explaining how scientific consensus is reached—and that this process corrects for the human errors in the long run.

The relevance of this to his point is obscure, but it gives him a chance to diss our paper and also drag Charles Darwin in as a racist and sexist, even though an accomplished one. Note that the link he gives below is to our merit paper, which he mischaracterizes as making the claim that science is “not subject to human influences”. He appears to be making an argument that judging science and scientists on merit is at odds with the view that scientists are human and flawed.  This is a false dichotomy that makes no sense.  Then he goes after Darwin and brings in race:

A raging debate has set in over whether the backgrounds and identities of scientists change the outcomes of research. One view is that objective truth is absolute and therefore not subject to human influences. “The science speaks for itself” is usually the mantra in this camp. But the history and philosophy of science argue strongly to the contrary. For example, Charles Darwin made major contributions to the most important idea in biology, but his book The Descent of Man contained many incorrect assertions about race and gender that reflected his adherence to prevalent social ideas of his time. [JAC: I’m curious about the “gender misconceptions”.] Thankfully, evolution didn’t become knowledge the day Darwin proposed it, and it was refined over the decades by many points of view. More recently, pulse oximeters that measure blood oxygen levels were found to be ineffective for dark skin because they were initially developed for white patients. These examples—and countless more in between—reveal how much work needs to be done to strengthen the scientific community and the public understanding of the process.

Darwin’s book On the Origin of Species also contains many errors not based on race and gender misconceptions. Of course Darwin was flawed, a man of his time who, by the way, happened to be more liberal on issues of race than most of his peers (he was an abolitionist). But that’s not even relevant to our claim that science should be judged on merit. We surely do not subscribe to the view that everything Darwin said was right because he was a good scientist: we judge what he accomplished on its merits. For a counterexample, he got genetics wrong, though it didn’t matter for most of his views. Only the last sentence above gives Thorp’s real claim: that better science can be done by diversifying the scientific community.

He expands on that in the next paragraph by ignoring a question that’s bloody obvious: if we are to diversify science by lowering the bar to entry (as Thorp apparently admitted during the debate), and downgrading merit to bring in more “diverse” people, then it’s obvious that our conventional ideas of “merit” must be given lower priority. Luana notes this above.  Thorp’s question, “How is diversity a threat to scientific rigor and the merit of discoveries? is in fact discussed in our paper. Our argument is not that diversity per se is a threat to merit, but that the drastic lowering of standards needed to attain full equity in science is a threat to merit. And we all recognize this. That is why, for example, grad schools are abandoning SATs as requirements for application, and why med schools do the same thing with MCAT tests.

What is likely is that diversity is promoted in science by people like Thorp primarily not primarily to improve science itself, but to make up for past wrongs done to members of minority communities (a form of reparations), to create better role models for underrepresented groups, to make scientists “look more like America,” or because diversity itself will create better science. In fact, the last point is what he maintains in the next paragraph:

A monolithic group of scientists will bring many of the same preconceived notions to their work. But a group of many backgrounds will bring different points of view that decrease the chance that one prevailing set of views will bias the outcome. This means that scientific consensus can be reached faster and with greater reliability. It also means that the applications and implications will be more just for all. How is this a threat to scientific rigor and the merit of discoveries? Unfortunately, we’re nowhere close to achieving these goals. Science has had enormous trouble building a workforce that reflects the public it serves. And now, numerous state governments are trying to make it more difficult, if not impossible, at the public universities in their states, and even within the scientific community, there are efforts to derail the idea that it matters who does science.

Talk about monoliths: the huge majority of scientists already share one viewpoint: the liberal democratic one.  In academia as a whole, surveys show that the ratio of Democrats to Republicans varies between five to one and fourteen to one! This is, of course, makes academia completely unrepresentative of America as a whole politically. Should Thorp be calling for this kind of diversity, too? No, because it’s not the right kind of diversity.

But the main flaw of this paper is threefold. First, Thorp gives no evidence that more diverse groups produce better science. My brief review of the data shows that there is some evidence that diverse groups can produce better results, but also that there is evidence in the other direction as well. (I am ignoring the very real possibility of ideologically-based publication bias here).  But Thorp’s claim above is not that, it is that you can have the same criteria of merit and also increase diversity. It’s the “you can have your cake and eat it too” argument. And this would hold if the increase in scientific progress accompanying a more diverse group of scientists more than compensates for the decrease in standards necessary to attain that diversity. And there is simply no evidence at all to support this. It’s telling that Thorp cites our paper and Paul’s column, but simply asserts that “different backgrounds. . . increase “scientific consensus”. If we really had good confidence that diversity actually increased the quality of science being done, then nobody would have a problem with boosting diversity!

Further, Thorp’s near-explicit claim that racial diversity will boost scientific quality is somewhat patronizing, as it assumes that Hispanics as a group, or blacks as a group, have an outlook on the world that will improve science more than other kinds of diversity: political diversity, viewpoint diversity, diversity in upbringing, whether one’s parents went to college or not, and so on. Looking at individual viewpoints and merit seems to me a better way than simply diversifying science to “look more like America.” If you want diverse viewpoints, find people with out-of-the-box viewpoints and hire them, but don’t assume that pigmentation or ethnicity automatically confers diverse scientific views that will push the field forward. The best way to push science forward is to give everyone equal opportunity and judge science and scientists on their merit. We haven’t yet accomplished the former, which is Task #1, but we can hold onto our standards of merit.

There is an empirically-based argument to be had about whether more diverse scientific groups produce better science. But we have no data to support that, and Thorp cites none. I can cite data on both sides, which means that there is no real consensus (some of the “pro-diversity” results, for instance, are based on mathematical simulations rather than real humans, while others are based on short-term problem-solving tasks in psychology laboratories). If the first claim proves to be true, then there’s another discussion to be had:  what types of diversity produce the best science? Do we need more Republicans? More people on the autism spectrum? Is it not possible that conservatives or people who are slightly autistic could ask questions just as different from mainstream scientists as, say, scientists of color? Why is ethnicity or gender the form of diversity claimed to best improve science? The answer, of course, is that we don’t know that, and the question itself is a diversion. Diversity is really being promoted for the same reasons it’s promoted in every field: as a form of reparations or to increase equality or equity.

Finally, even if diversity of one type or another advances science, we need to show that the erosion of the meritocracy required to make the field more diverse is more than compensated for by the net increase of scientific progress produced by having more diverse scientists. We aren’t even close to knowing that, and I doubt, given the kind of data we’d need to show it, that we ever will.

So Thorp is just blowing smoke, and also bringing in an irrelevant claim that somehow our failure to see scientists as humans has drastically hurt science. That, in fact, is how he ends his post:

Scientists should embrace their humanity rather than pretending that they are a bunch of automatons who instantly reach perfectly objective conclusions. That will be more work both in terms of ensuring that science represents that humanity and in explaining how it all works to the public. But in return, society will get better and more just science, and it will allow scientists to immerse themselves in the glorious, messy process of always striving for a greater understanding of the truth.

Here he’s arguing against something that no scientist maintains. Maybe the layperson thinks that scientists are a bunch of automatons, but we scientists know better. But most important, Thorp never explains how our recognizing that we are fallible humans (which we already know!) will suddenly boost the progress of science.

What bothers me most is that the editor who controls what may be the most powerful and important science journal in the world is incapable of making a coherent argument, or laying out what data would be needed to support his claims. He is very big on assertions and very short on facts. Is that the kind of science editor we want?