The “Systemic Racism” MIT Debate – Notes and Observations

On November 2^nd, 2023, an MIT AFSA debate took place.  The proposition was: “Resolved, that STEM is systemically racist.” A video recording of the debate is available here.

The negative team, answering “no” to the proposition, was Dr. Erec Smith, Associate Professor of Rhetoric at York College of Pennsylvania and President and Co-founder of Free Black Thought, and myself.  On the affirmative side were Dr. Chad Womack, Vice President of National STEM Programs at the United Negro College Fund (UNCF) and Dr. Jaret Riddick, Senior Fellow at the Center for Security and Emerging Technology at Georgetown University, who formerly was Principal Director for Autonomy in the Office of the Undersecretary of Defense for Research and Engineering at the Pentagon. Dr. Womack coauthored this paper in Cell, and was a guest editor of a series of special issues in Nature on racism in science.

Below are my impressions of the debate as well as my complete eight-minute opening remarks (to fit within the allotted time, I abbreviated several sentences during the debate).

My impressions

The purpose of the debate was fulfilled.  We were able to have a respectful and productive discussion about a very controversial topic.  The debate appeared to engage the audience: public questions continued until after 10 pm—and could have gone on for much longer.  In the end, both sides came to agreement on many of the issues debated.  For instance, both teams agreed that inequities in STEM representation start in K12, long before students enter STEM fields, and agreed that some “anti-racist” initiatives, such as “un-grading” (self-grading), citational justice, and other initiatives that decrease rigor with the intent of “equalizing” grades, are counterproductive, likely increasing inequities.  As Erec Smith put it, these initiatives are in fact quite racist, as their rationale seems to be that Latino or Black children need special accommodations because they learn “differently” and that intellectual rigor and debate are “white values.”

A Proceedings of National Academy of Sciences NEXUS (PNAS Nexus) paper was repeatedly cited by the affirmative team.  They used the results of this paper, which allegedly shows that Blacks and Latinos who initially enroll in STEM majors ultimately graduate with STEM degrees at much lower rates than do whites (even after controlling for grades), as evidence that systemic racism is indeed pervasive in STEM.  But even if this were the case, there are many reasons Latinos and Blacks could be switching away from STEM, with many going into “Identity Studies” majors.  For one thing, Latinos and Blacks will be the groups most affected by the victimhood ideology that is constantly pushed onto them – such as the idea that virtually any interaction is a “microaggression”.  Further, underrepresented minorities would be negatively affected by narratives that insist that they do not belong in these disciplines because STEM is characterized as “white.”  Finally, there are other aspects that were not controlled for in the study mentioned, such as socio-economic class and culture. Over my years of teaching, I had several Latino students who were interested in continuing their studies in ecology, but whose parents were not supportive, claiming that a science not related to medicine is not worth studying.  I experienced this myself when at 11 years old: my father derided me and ordered me to go to my room when I announced that I wanted to be an ecologist. 

After the debate I looked up the paper that was so heavily cited by the affirmative team, only to find that the conclusions are not that straightforward.  The senior author is Chad Topaz – a Williams College professor previously part of the Math Department, but now in the Humanities Department. Prof. Topaz was behind the cancellation of Prof. Abigail Thompson, due to her criticism of the use of diversity statements in job hires.  The ensuing cancellation resulted in many letters both in favor and against Prof. Thompson’s argument.  Following the controversy, Professor Topaz published a paper on the ethnic and gender "diversity" of signatories of both his own letter (against Prof. Thompson) and the letters in defense of Prof. Thompson, initiated by other mathematicians. It is interesting to note that Prof. Topaz’s organization used to receive donations to help others write diversity statements for academic positions.

The authors of the PNAS Nexus paper concluded that, after controlling for academic preparation in high school and stated intended major, underrepresented minorities (URM = Latinos and Blacks), still drop out at a higher rate than do Whites. However, the way the authors controlled for grades was not by using grade point averages, but by a binary trait: grade point average above or below C-minus (DWF or above). The authors also failed to use most of the colleges available in their dataset, winding up with a small sample of only six R1 research universities.  I find it interesting that they used this method to control for grades, as this apparently puts a large number of disparately achieving people into the same group (from A+ to C-!).  It is also puzzling as to why the authors include the “W” (“withdrew”) among D and F grades – after all, a “W” can result from all kinds of reasons, from illness to surgeries to a death in the family.  It is further notable that there were no controls for socioeconomic status, family education, and other cultural factors that are likely to make a difference in what degree a student pursues. While it is possible that lower STEM graduation rates of Blacks and Latinos would still remain after controlling for grade point average (instead of DWF grades), I would like to see these analyses done.

Despite the vigorous exchange of views, my own position remains largely unchanged.  The huge differences in educational attainment and STEM participation between Blacks and Latinos and Whites and Asians, start before college, so cannot be due to systemic racism acting after one enters college. These differences are likely caused by many factors, but are probably due mainly to differences in socioeconomic status and culture. They will not be ameliorated (and will likely be made more severe) by recent “anti-racist” initiatives such as dropping meritocratic standards as a way to “equalize” achievement. 

I thank the MFSA for organizing this debate and I hope it will generate discussion in many other venues. 

My 8-minute opening remarks

The core of this debate depends on definitions: especially what we mean by “STEM” and “Systemic racism”. In fact, I think if we can come up with agreed upon definitions, all of us would likely agree on much of what is at stake here. 

Let’s start with an easy one: “racism” or “sexism”, which I take to mean discrimination against individuals because of their race, sex, or other immutable traits. 

But what do we mean by “systemic racism”?  

Systemic racism is more than “a few racists”– it could denote either rules codified to exclude minorities or at least a general atmosphere of discrimination shared by others, but not codified or made explicit.  Both of these would cause a systematic exclusion of people of color. 

STEM is trickier to define.  Are we talking about individual researchers?  Are we talking about our research institutions?  Are we talking about particular scientific results?  Or are we talking about the scientific method itself? 

There is no argument that the scientific toolkit of investigation, controls, replication, prediction, blind studies, experimentation, statistical tests and modelling—cannot by itself be racist.  After all, science, while developed mostly in the Western world, has now been adopted by many cultures and has brought huge benefits to nearly everyone.  For instance, the way “science” is done in Brazil is the same as in the US – I had no trouble at all moving the “science” that I learned in my home town of Rio de Janeiro to Cornell University. 

Scientific results and interpretation are constantly re-inspected – in science nobody has the last word.  Science is self-correcting and the racist “science” from the past - think of racial hierarchies - is now considered a joke.  Science aims to see the world as it is – and this goal itself cannot be racist.  

Clearly individual researchers can be racist, but that doesn’t seem to be what people mean by “systemic” – unless the majority of scientists are still racist and jointly push for discrimination. In my many years participating in science in various countries in the world, I have not observed anything like coordinated mass discrimination. In fact current data demonstrates that it is easier to be hired as a woman than as a man; and the excess of Whites claiming to be Native Americans or Blacks suggests the perceived advantages one gains as a person of color (two examples are Margaret Noodin, Milwaukee professor of American Indian Studies who passed as a Native American, and, Jessica Krug a George Washington University professor of African and Latin American studies who passed as Black).  In, fact, if there is any system in place today, it is the DEI system – which enforces norms of “anti-racism” and openly demands preference for people of color through ideological litmus tests such as requiring DEI statements in faculty hires and even in graduate-student admissions.   These systems, and further commitments to “diversity”, are also abundant in most prestigious scientific publications, such as Nature and Science.

Now that we have defined some terms, let’s move on to what we see in STEM today.  But before that, let me state that there is absolutely no question that racism was prevalent in the recent past and it was very difficult for women and racial minorities in the USA or Europe to participate in the scientific enterprise – This was for two reasons: a lack of opportunity when younger and a direct result of existing racism and sexism preventing advancement. In fact, it should be obvious that most scientists in the past held racist and sexist views: Views that often tainted their own scientific investigation (think about the old trope that Blacks feel less pain), and even how they applied the concept of merit to those who were different from themselves (think of Jocelyn Burnell, who discovered pulsars, but was not recognized when her advisor got the Nobel Prize instead of her). 

There should also be no question that times have changed dramatically since the 1960s.  Science is now much more open to those who can prove themselves meritorious.  The racists and sexists who still exist in STEM have almost no power today and do nothing to impede the advancement of people of color and women in today’s STEM fields.

Yet despite massive cultural changes that have decreased racism and sexism, there is also no question that we see a huge underrepresentation of Blacks and Latinos in science.  As an example, at MIT only 4% of professors are Black and 5% are Latinos (compared to 13% of Blacks and 18% Latinos in the overall population).  In the US as a whole, the number is similarly low, with 7% Black and 6% Latinos professors. 

The situation for women has changed and now some 40% of American bachelor’s degrees in STEM go to women.  However, while women are abundant as professors and students in fields like psychology, in other areas like physics they are quite scarce. At MIT, 35% biology professors are female, but that drops to just 15% for physics.

These numerical inequities have served for many as prima facie proof of racism and sexism. But the inequities can be explained by lots of factors that might not involve either racism or sexism. In fact, simply referring to "systemic racism" and looking for quick remedies for numerical inequity, stands in the way of understanding the real causes of disparities. We should use scientific tools to explore these questions, because if we come to the wrong conclusions about causation, we will not be able to find effective solutions.  

Why are there so few male elementary-school teachers or nurses?  Why do females constitute 85% of gynecologists but only 9% of urologists? Why are 70% of NFL players black?  Why are so few women firefighters?  Are all of these disparities caused by racism and sexism? Or do they result from differences in biology, group preferences, or culture?  The same questions apply to the disparities we see in STEM. 

What if the conditions that lead to inequities in STEM start so early that by the time students are applying to graduate school, these disparities are already in place? According to the National Assessment of Educational Progress (NAEP), in 2022 only 8% of Black pupils in 12^th grade were proficient in Math, and this number was only slightly higher for Hispanics (11%).  In contrast, 32% of Whites and 52% of Asians fell into the proficient category, with other areas of knowledge showing similar disparities.  Assuming that proficiency in math is required for college and graduate work in STEM, we can adjust for demographics. When we do this, we find that 8% of the 13% Blacks will be expected to be ready for advanced scientific work, 52% of the 6% Asians and so on.  Bring these adjusted percentages back to 100%, and we would expect that “proficient” students would be about 70% White, 12% Hispanic, 13% Asians and 4% Black. And, in fact, this is almost exactly the proportions we observe at MIT graduate school; with Whites being the only group a bit underrepresented in relation to expectations – constituting 60% rather than 70% of graduate students. The rest--12% Hispanics and 4% Blacks – are exactly the proportions expected from their adjusted proficiency tested at the end of high school.  It is thus clear that the “pipeline problem” starts long before individuals decide to enter STEM. Its cause is not a result of systemic racism in science but differences in proficiency that may begin even before students finish elementary school.

In conclusion, of course racists are still with us (albeit at much lower number than in the past), and of course racism and past racism in particular might still be responsible for the inequality of opportunity that starts at birth.  But this kind of individual racism is so rare that it cannot play a substantial role in generating the inequities seen in today’s STEM.  

In the end, you must ask yourself:  do you know any qualified Black scientist today who, because of discrimination—and not lack of merit—could not enter STEM simply because he or she was Black?