How do you know what you know?

from Don Taylor’s RECKONING BLOG at https://bit.ly/3uqoWUN on February 24, 2021 at 12:45PM

Peder Zane has a piece in the WSJ based on an interview with James B. Duke Professor Emeritus of Psychology, Neuroscience and Biology, James Staddon, criticizing Duke University’s “Anti-Racism” efforts, couched as defense of Science. The essence of the interview is that Professor Staddon is a scientist, and that many Duke faculty are not, especially those in the Social Sciences, because they do not follow the semantic reductionism of Randomized Control Trials (RCT)s or adhere to the internal validity conventions of animal-based model research.

“weak science competes with activist political tendencies around fraught issues of race, class and gender.”

This leads to the conclusion of a University that has lost its way and no longer seeks truth alone, but is instead a tool of political indoctrination.

The span and scope of the material covered in the piece is centuries long, and the disciplines involved touch nearly every corner of Duke, yet Professor Staddon and Mr. Zane seem fully sure of their conclusions. However, I would like to grant Professor Staddon the benefit of the doubt and assume he is interested in answers to the questions he posed about Duke and Race in the WSJ piece. It is in the spirit of answering a colleague that I offer this post.

“What are these ‘systems of racism and inequality?’ How have they affected Duke and how is Duke involved in them?”

From 1931-1964, Duke University included Racial Covenants (see the pdf download below) in the deeds of 300 homesites in the Duke Forest neighborhood that it sold to faculty and staff that remain today, even though legally unenforceable for decades. These covenants state that people who were not White could not stay overnight in the house unless they were servants. I once owned one of those homes, built in 1959, and asked about it at closing in 2006, and my lawyer said it was unenforceable but difficult to change given how the covenants had been written. Thus, Duke: (1) imposed Racial segregation in robust form to make Durham palatable for White faculty; (2) did not fully tell this story in a transparent manner; people (like me) knew about it for a long time; (3) these covenants helped define the residential segregation patterns seen in Durham today that influence things like school assignment, and public services, including policing; (4) residents of Durham, especially those who are Black, understand that at the heart of Racial covenants is an idea–some people are worth more than others, and the persistence of this idea is how “systems of racism and inequality” still affect Duke.

One more example. Duke hospital and the University’s physician practice plan had segregated floors and clinics from the 1930s until the early 1960s. They were replaced by the nomenclature of “private v. public” even before segregated medical care became illegal, and with the advent of Medicare and Medicaid in 1966, huge inflows of money began as health insurance coverage surged, helping make Duke University what it is today. The full story of how this shift in language nevertheless lead to continued segregation (when in doubt, Black patients were “Public”, White patients were “Private”) is just now in the process of being fully told, in part due to the fact that Board of Trustees records are confidential for 50 years. There is a straight line from this history—deeply known through experience and mostly untold—to the current reticence at receiving the COVID19 vaccine by some Duke employees and Durham residents who understandably doubt that Duke has their best interests at heart. This affects us all as we seek herd immunity.

These are just two “systems of racism and inequality” that Duke University built to segregate both housing and health care based on Race, and those systems influence our world today, at Duke, in Durham and in the lives of all the students we have taught and trained over the years. Naming and addressing these realities and blind spots is key if we as a University claim to have, or aspire to a more equitable culture of scientific excellence (the bold is my goal for Duke, and I understand the broad efforts to address Structural or Systemic Racism at Duke to be part of making that goal a reality).

How do systems of racism and inequality affect health? Broadening the discussion beyond simply answering Professor Staddon’s questions with the two examples offered above, the vision of the life of the mind described in Mr. Zane’s piece is too narrow to address many of the most important problems of our day. While animal model experimentation is a crucial part of the cannon of knowledge produced by a Research University, this approach alone cannot ask, nor answer all the important questions. A far more difficult and nuanced approach is called for to understand cause and effect in areas that cannot be studied in reductionist fashion via placebo and randomization, or measured only at the cellular level. This does make it difficult to find agreement about “how do you know what you know?” an important question raised in Mr. Zane’s piece by Professor Staddon. Let me try and explain what I mean.

The 2009 Nobel Prize in Physiology and Medicine was awarded for advances in understanding Telomeres, caps on DNA strands that provide protection during chromosome replication, and whose length provides a measure of cellular aging. An emerging body of research suggests that cellular aging is more rapid for Blacks as compared to Whites, even after controlling for other factors that influence Telomere length, such as poverty, obesity, smoking and exercise. More work is needed to calibrate the practical impact on life span and to so determine what proportion of a Black/White mortality gap could be explained by differential cellular aging. Science builds upon on Science. The 1983 Nobel Prize in Physiology and Medicine was awarded to Barbara McClintock, for her discovery in the 1930s of “mobile genetic elements” identified first in maize and then fruit flies. This forgotten finding became a building block for the 2009 Nobel Prize that shows the progression of findings from plant, to animal model, to human discovery and finally to the potential to intervene in the lives of human beings–the march of Science.

However, others were working as well, motivated not by an intrinsic interest in human biology, but instead by an explicit focus on explaining why Black/White mortality differences were so large and persistent. A finding that cellular aging is more rapid for Blacks versus Whites net of other factors, is consistent with an interpretation that Racism causes negative health impacts, providing a link between structural factors such as those noted above and individual health, most likely via the bodies stress response as measured by allostatic load. My Duke colleague, Susan B. King Distinguished Professor Emeritus of Public Policy Sherman James first proposed John Henryism, a hypothesis to explain this Black/White mortality gap nearly 40 ago, based on the idea that striving against the headwinds of being Black lead to individual harm, with numerous mechanisms hypothesized as potential causal links (see pdfs linked at the end). Professor Arline Geronimus followed with the Weathering Hypothesis in the early 1990s to explain Race by Gender mortality differentials, and this work has been tested using allostatic load and cellular aging techniques in addition to being applied to other outcomes (pdfs at the end). The 2009 Nobel Prize in Medicine provided a means of using biological evidence in the form of Telemores, to study the impact of structural Racism on individual health. The march of Science.

The use of Telomeres to document differential aging by Race remains an emerging field that is limited by the samples available for study (selection bias is always an important factor when interpreting research, whether a RCT or not, and there are tradeoffs between internal and external validity). More research, as always, is needed and understanding Black/White mortality differences affects us all, because our assumptions about the causes influence what we think about one another and therefore how we respond, individually or as a society via public policy. Life expectancy at birth is 4.6 years less for Black males as compared to White males; and 1.8 years shorter given survival to age 65 (red underline males; yellow highlight females–table is in this post).

The best that the modern Research University can offer is an amalgam of scholars doing what they do in the manner that they do it, and disputes both within and across disciplines are legion. It is difficult to synthesize scholarship across disciplines, time and methods. Maize, mice, humans, telomeres, bench to bedside translation, history, public policy, business, art, literature and ethnography. This is who we are collectively as a modern research University, and just as Barbara McClintock’s discovery gained dust for decades until she won the Nobel Prize in 1983, human knowledge progresses in a non-linear fashion, and you never know when a unique insight will provide the missing piece of the puzzle.

Answering whether Racism or “living while Black” is an independent predictor of health, or simply a proxy for other factors cannot be investigated in an experimental manner. We cannot go back and randomize some to live under Structural Racism and others to not, or to run 1,000 experiments of “Reconstruction” and identify how different courses of history over the past 150 years might change the distribution of illness and longevity by Race that we see today. The 1862 Homestead Act, in force and revised numerous times until its repeal in 1972 distributed 10 percent of the land in the United States to individuals. Had this program not had a Century long history of discriminatory practices against freed Slaves and Blacks, we would undoubtedly observe a different wealth distribution and life expectancy spread by Race. Counterfactual thinking is hard, and requires painstaking effort to piece together what different disciplines have to say.  

How Do You Know What You Know? Zane’s piece describes Professor Staddon as simply wanting to ask “how do you know what you know?” and having no interest in politics. I agree that this question is central to the life of the mind and the University, but the Harold Lasswell definition of politics still rings true from graduate school–“Who gets What, When and How?” Nothing is apolitical, including, or perhaps especially, assuming that only those things that can be studied via experimental or reductionist evidence are knowable, or worth knowing. Interestingly, a book that I wrote with Frank Sloan and others (The Price of Smoking MIT Press, 2004) apparently played a role in Professor Staddon’s “tipping point” to concluding the University has been given over to political correctness , based on the quote in Mr. Zane’s piece of him mis-interpreting one of our book’s findings:

Still, he says, “my personal tipping point” regarding the uses and misuse of science occurred some two decades later, “when I found out that despite massive publicity to the contrary, smoking has no public cost.”

This is not what we found. This post from 2011 explains what I mean—cigarette taxes did on average cover the external costs of smoking, but did not cover quasi-external ones, and we explicitly assigned a dollar value of $0 to the intangible cost of life years lost. We did this because the range of estimates in the literature differed by orders of magnitude, thus allowing readers to use the per pack estimate they believed to best represent the intangible costs of smoking mortality that we left aside in order to painstakingly highlight the economic costs and their distribution. Many have made a similar assumption about the book, and there is much nuance of interpretation, and very little of this work is based on experimentation. However, our work was obsessed with identifying the best possible counterfactual given that experimentation is uncommon or impossible in the study of Smoking outside of pharmaceutical work and social-psychology based perceptions of risk, and our concept of the non-smoking smoker, estimated via Smoking Life Tables, and the identification of “quasi-external” costs, or those imposed by smokers on their families are the most important theoretical and methodological innovations of the book. If such research is not Science then most of what impacts human beings today cannot be studied by Science.

I think our book is great, and you should buy 20 copies. However, if the distributional analyses of one book (no matter how good!) on the most heavily studied topic of the past 80 years (the impact of Cigarette Smoking on humans) could send Professor Staddon over the top to his conclusion that Universities have given into political correctness and lost our way, then to paraphrase the Eagles, that had to be where he already knew how to go.

PDFs of selected relevant works mentioned above, but not readily available via hyperlink are below