Research

Note

In November 2024 I realized I hadn’t updated this in like five years. The text below comes from my application for tenure, which I submitted in June 2022. More recent publications will show up chronologically on my CV and ORCid profile.

Since 2011, I have published 21 peer reviewed articles, along with 16 other scholarly products that were not peer reviewed, such as book chapters, commentaries, and book reviews. Of the 21 peer reviewed articles, 15 were single-authored; I was lead or co-lead (first author or equivalent) of the other six.1

John Dewey described the relationship between science and society in this way:

Scientific subject-matter and procedures grow out of the direct problems and methods of common sense, of practical uses and enjoyments, and … react into the latter in a way that enormously refines, expands and liberates the contents and the agencies at the disposal of common sense. (Logic, The Theory of Inquiry, 71-2)

On Dewey’s model, the traditional ideal of socially disengaged “pure science” is misleading, both descriptively and normatively. Scientific research questions trace their roots to the concerns and problems of the broader social context; and when things are going well scientific findings make useful contributions to those concerns and problems. I believe a similar model is applicable to the history, philosophy, and social studies of science (HPSTS). Our research grows out of the problems and methods of both scientific practice and everyday “common sense,” and when done well contributes to both.

Indeed, Dewey’s model suggests that the distinction between scientific practices and their social context is artificial and fuzzy. Scientific questions and methods are somewhat refined versions of those of “common sense,” not a radically different kind of activity. Similarly, for me, the questions and methods of HPSTS can’t be neatly distinguished from the questions and methods of scientific research, or other activities in our broader social context.

Consequently, my research has increasingly ranged beyond the bounds of disciplinary philosophy, but has remained socially relevant. Combined with my work experience in science policy and data science, this means that my research doesn’t fit neatly into a single disciplinary research program. Instead, in this section of my statement, I organize my research into four themes or areas: feminist philosophy; the aims approach to science, values, and policy; public scientific controversies; and the use of computational methods.

Feminist philosophy

Most of my work in disciplinary philosophy of science is in the subfield we call science, values, and policy. Today, this subfield is burgeoning, with numerous sessions at meetings of the Philosophy of Science Association (the primary professional association for philosophers of science) and articles appearing regularly in our major journals. But when I finished my dissertation in 2012, science, values, and policy was not yet a recognized subfield. The publication of philosopher Heather Douglas’ Science, Policy, and the Value-Free Ideal (2009) had just started to mainstream criticism and rejection of the ideal of value-free science. Before Douglas’ book, feminist philosophers of science had been the most prominent critics of the value-free ideal. The Science Wars — exemplified by Gross and Levitt’s Higher Superstition (1994) and the Sokal hoax (1996) — had cooled off around 2000. But feminist analyses of the relationship between science and society were still associated (incorrectly) with “postmodernism” and the methodological relativism of the Strong Programme in the Sociology of Scientific Knowledge. As reflected in its title, Scientific Practices and their Social Context (2012), my dissertation was directly interested in the relationship between science and society. Feminist philosophy of science provided one of the two major conceptual frameworks that I used in my dissertation — the other was the conception of practice developed by the Thomist-Marxist ethicist Alasdair MacIntyre — and some of my first publications were direct contributions to feminist philosophy.

“Is Longino’s Conception of Objectivity Feminist?” (2011) developed and considered responses to a critique of feminist philosopher Helen Longino’s influential social conception of objectivity. I emphasize that I did not originate this critique; Natalia Baeza, another graduate student at Notre Dame at the time, presented it in a seminar we took with Janet Kourany. I developed the critique more formally, and showed that my initial response (in seminar) was not successful.2 Briefly, Longino’s account of objectivity seems to require that scientific communities actively cultivate not just feminist and antiracist scientists as Longino intended, but also antifeminist and racist scientists. I called this “the Nazi problem”: Longino’s account seems to require actively cultivating Nazi scientists. Given the recent resurgence of eugenics, race science, and fascist political movements in the US, today I think that the Nazi problem is more than merely theoretical.

“On Okin’s Critique of Libertarianism” (2015) systematically addresses every published response to feminist political philosopher Susan Moller Okin’s critique of (political) libertarianism in Justice, Gender, and the Family. Okin argues that Robert Nozick’s libertarian account of self-ownership contradicts his account of the acquisition of property through labor. On its face, this is a devastating critique of the most common strain of libertarianism, which regards the account of property as logically entailed by self-ownership. If Okin is right, self-ownership is self-contradictory. To present her argument simply, women (pregnant people) engage in labor to make people; so either mothers own their children or self-ownership requires the forced transfer of property. I present Okin’s argument in a deductively valid form, classify the published responses in terms of the premises they challenge, and show that all the substantive responses to Okin face the same dilemma: preserving the principle of self-ownership requires forced transfer of property.

While feminist philosophy has had a lasting influence on my work, and I regularly engage with feminist philosophers in other publications, these are the only two publications that directly contribute to feminist philosophy.

The aims approach

The second major theme or area of my work is development of what is called the “aims approach” to science, values, and policy. This approach argues that scientific research typically has both practical as well as epistemic aims, and that it’s legitimate for both kinds of aims to shape the conduct of research, including research design, data collection, and the evaluation of hypotheses. For example, conservation biology has the aim of protecting ecological communities, and public health has the aim of protecting the health of human communities. Recognizing these practical aims does not mean sacrificing the pursuit of knowledge, but instead seeing knowledge as the primary means by which researchers promote the practical aims of their field. But this means successful science needs to be both epistemically and practically valuable.

During and shortly after finishing my dissertation, I was primarily interested in the philosophical project of developing a conceptual framework for my specific version of the aims approach. Publications engaged in this conceptual development include “On the Ideal of Autonomous Science” (2011), my dissertation Scientific Practices and Their Social Context (2012), and “A New Direction for Science and Values” (2014). A few years after I finished my PhD, the historian of economics Thomas Stapleford approached me about collaborating on an historiographical essay, presenting ideas from my dissertation as a methodological framework for the history of science. This collaboration led to “The Virtues of Scientific Practice: MacIntyre, Virtue Ethics, and the Historiography of Science” (2016) and later “Seeing Science as a Communal Practice: MacIntyre, Virtue Ethics, and the Study of Science” (2021).

In particular, “A New Direction for Science and Values” (2014) is my most highly-cited publication, as one of the standard references for the aims approach.3 The paper presents a puzzle: examples from fields such as primatology and archaeology show that feminist values can have a positive influence on research; but cases such as the Vioxx scandal show that industry values can have a negative influence. Even when we reject the value-free ideal, we need a way to distinguish legitimate from illegitimate influences of values in science. I suggest reflecting on the aims of science allows us to resolve the puzzle. The “new direction” of the title is deep engagement with ethics and social and political philosophy, which can provide philosophers of science with more sophisticated conceptual tools for thinking about the aims and flourishing of social groups.

My time in science policy, 2015-17, make me much more interested in disseminating my work beyond disciplinary philosophy journals. Two co-authored papers have presented ideas from my version of the aims approach for research assessors and environmental scientists and policymakers, respectively. “Impacting Capabilities: A Conceptual Framework for the Social Value of Research” (2018) was written with digital humanist Carl Stahmer and MacKenzie Smith, University Librarian at UC Davis. We argue that political philosopher Martha Nussbaum’s list of “central human capabilities” can provide a rich framework for articulating the practical aims (“broader impacts”) of research, and show how text mining methods could be deployed in a research assessment context to efficiently identify research intended to promote various practical aims. This paper was published in the research assessment journal Frontiers in Research Metrics and Analytics. “Legitimizing Values in Regulatory Science” (2019) was written with philosopher of science Manuela Fernández Pinto and published in Environmental Health Perspectives, a major environmental public health journal. We make a novel argument that the value-free ideal is self-defeating in the context of regulatory science and, building on the idea that regulatory science has the primary aim of protecting human health and the environment, discuss institutional forms such as participatory research and participatory governance that, we argue, can legitimize the explicit appeal to values in regulatory science.

Public scientific controversies

The third major theme of my research is the analysis of public scientific controversies. This theme emerged directly out of community engagement: During my last few years of graduate school, I became involved with a local food co-op in South Bend, Indiana. As one might expect, almost all the members of the co-op had strong environmentalist convictions, and many of the members were strongly opposed to genetically modified (GM) crops. However, at least some members — primarily faculty and graduate students associated with Notre Dame’s biology programs — strongly supported GM crops. My participation in the co-op led to a series of papers examining several aspects of the GM crops controversy more broadly, including “Epistemological Depth in a GM Crops Controversy” (2015), “GMOs: Non-Health Issues” (2016, an encyclopedia article written with philosopher of science Roberta Millstein), and “Genetically Modified Crops, Inclusion, and Democracy” (2017). These papers argue that concerns about the safety and productivity of GM crops are deeply entangled with concerns about the economic power of multinational agricultural biotechnology companies and the industrial food system as a whole. But, in the US, the regulatory process for approving GM crops is based exclusively on safety assessments. The technocratic, scientific controversies over yields and safety are thereby proxies for much deeper political disagreements, and so scientific evidence is unable to resolve the controversy.

In “Scientific Controversies as Proxy Politics” (2017), I generalized these insights from the GM crops controversy for readers of the science policy magazine Issues in Science and Technology. I argued that the scientific controversies over vaccines and climate change were likewise proxies for deeper disagreements, about medical paternalism and the economic power of the fossil fuels industry respectively. When this article was published, I participated in a panel discussion with science journalist Keith Kloor at the Arizona State University “embassy” in Washington, DC. While the article was in press in fall 2016, I discussed similar ideas in an invited public lecture at Hofstra University, one of a series of events they held before the first presidential election debate. And a few months after the article was published the Society for the Psychological Study of Social Issues invited me to give a public lecture on the same topic.

As I was developing the “proxy politics” framework, I had made the move from academia to science policy. During 2015-16 I was a AAAS Science & Technology Policy Fellow with the US Environmental Protection Agency’s Chemical Safety for Sustainability Program (CSS); then I spent 2016-17 with the National Robotics Initiative at the National Science Foundation (NRI). Both of my host offices were specifically interested in what I could offer as an analyst of public scientific controversies. I wrote “Inductive Risk and Regulatory Toxicology: A Comment on de Melo-Martín and Intemann” (2018) for CSS, and “The Safety of Autonomous Vehicles: Lessons from Philosophy of Science” (2018) for NRI.4 Both pieces deployed the “proxy politics” framework. In the former, I showed that a disagreement between US EPA and the Natural Resources Defense Council (NRDC) over a new chemical screening technique came down to different ways of estimating false negative rates, which in turn corresponded to different ways of prioritizing the protection of human health and the environment. The latter piece identifies one way in which safety comparisons between human drivers and autonomous vehicles could become politicized.

My time at EPA made me especially interested in controversies surrounding environmental public health. “Legitimizing Values in Regulatory Science” (2019, with Fernández Pinto, discussed above) focuses on such controversies. In May 2020 I submitted a public comment on a proposed US EPA rule, then called Strengthening Transparency in Regulatory Science, that would have imposed a strong open science requirement on environmental public health research. Strengthening Transparency explicitly appealed to the ongoing replication crisis to justify the rule. Drawing on my experience working as a data scientist (discussed further in “Computational Methods” below), I emphasized a distinction between computational reproducibility and experimental replicability, then argued that the replication crisis was not primarily a reproducibility crisis; that there is no evidence of a replication crisis in environmental epidemiology; and that, even if there were, the kind of open science requirement imposed by the rule would be incapable of addressing a replication crisis. After attempting to publish this as a commentary in a major environmental public health journal,5 it was published as “Open science, the replication crisis, and environmental public health” (2021) in the research ethics/policy journal Accountability in Research.

A pair of pieces, “The P value plot does not provide evidence against air pollution hazards” (2022, peer reviewed journal article in Environmental Epidemiology) and “Young’s P-Value Plot as an Agnogenic Technique” (2022, blog post on Social Epistemology Review and Reply Collective) offers statistical and philosophical responses, respectively, to critiques of air pollution epidemiology made by the statistician S. Stanley Young. Using a graphical method he calls “the p-value plot,” Young alleges that this area of research is marred by publication bias (in which researchers do not publish undesirable results), p-hacking (in which researchers run various alternative analyses until they achieve statistical significance), and a failure to account for statistical heterogeneity (the possibility of different statistical effects in different subpopulations). My statistical paper uses a computer simulation to show that the p-value plot as used by Young cannot provide evidence to support his criticisms. And the philosophical blog post argues that the p-vaue plot is an “agnogenic technique,” that is, a technique for producing ignorance (about the hazards of air pollution) rather than knowledge. Beyond agnogenic features of the plot on its own, I discuss how I had difficulty publishing this paper, because reviewers at multiple scientific journals considered direct technical criticism to be an ad hominem attack on Young. In this way, the norm against ad hominem attacks — which normally has a knowledge-promoting function — actually frustrated the production of knowledge.

My most recent publication, “When Virtues are Vices: ‘Anti-Science’ Epistemic Values in Environmental Politics” (2022), combines the two themes of the aims approach and public scientific controversies. As in other work I start with a puzzle related to a controversy: the environmental policies of the Trump administration were widely criticized as “anti-science”; but in a few cases the primary arguments for these policies appealed to widely-shared epistemic (knowledge-promoting) values, such as avoiding error and promoting transparency through open science. How can policies that promote epistemic values be “anti-science”? To resolve this puzzle, I draw on axiology (the theory of value) and virtue ethics to provide a nuanced account of the relationship between epistemic and practical aims of science. This account allows me to explain how scientific knowledge can be both intrinsically valuable (valuable for its own sake) and instrumentally valuable (valuable for other, further goals), and argue that when there is an apparent conflict between immediate epistemic aims and ultimate practical ones, epistemic aims should be reconceptualized to bring them into alignment with the practical aims. Because the “anti-science” policies would have frustrated the pursuit of the practical aims of fields like environmental epidemiology (protecting human health and the environment), these policies are indeed “anti-science.”

Computational methods

The final major theme of my research is the use of computational methods, especially bibliometrics, network analysis, and text mining. During my time in science policy, I was surrounded by natural and biomedical scientists and engineers. I was often the only person in the room with any formal education in the humanities at all, and on more than one occasion when I introduced myself as a “philosopher of science” people had no idea what this combination of words could possibly mean. While my colleagues were supportive, I felt a subtle but pervasive pressure to justify my expertise.

I decided that bibliometrics6 provided a natural entry point to quantitative empirical research, since its methods and research questions overlapped with science and technology studies (STS). My first fully empirical publication, “Bibliometrics for Social Validation” (2016), used a citation network to examine the uptake of US EPA’s toxicology research by the broader scientific community. “The National Robotics Initiative: A Five-Year Retrospective” (2019, coauthored with Reid Simmons, then then NRI program manager at NSF) and “Network analysis to evaluate the impact of research funding on research community consolidation” (2019, coauthored with David Coil, Carl Stahmer, and Jonathan Eisen) used coauthor networks to analyze how particular research programs had supported novel interdisciplinary collaborations. “Productivity and Interdisciplinary Impacts of Organized Research Units” (2021) likewise examined interdisciplinarity, here among the Organized Research Units (ORUs) at UC Davis, but using text mining methods rather than network analysis. All of these four papers were done, as it were, “for hire”: in each case I was a temporary visitor (policy fellow or postdoc) and conducted these projects for my host office or research center.

Recently I’ve begun to apply these computationa methods to my own research interests, especially public scientific controversies. The simulation study of the p-value plot — relevant to controversies over air pollution — was discussed above. “Census Demographics and Chlorpyrifos Use in California’s Central Valley, 2011–15: A Distributional Environmental Justice Analysis” (2020, published in the International Journal of Environmental Research and Public Health) began as a project for a graduate methods course on spatial regression that I audited at UC Davis. The paper uses public data from California’s Department of Pesticide Regulation (Cal DPR) and spatial regression methods to analyze racial-ethnic disparities in the local use of chlorpyrifos, a highly controversial pesticide and known neurotoxicant that until recently was heavily used in the Central Valley. My major finding was that a 10-point increase in the Hispanic proportion of a census tract was associated with an estimated 5%-40% increase in local chlorpyrifos use.

The Computational History of Philosophy of Science Dataset (CompHOPOS) was an effort to assemble a comprehensive dataset of professional philosophy of science, including both metadata and fulltext of articles/chapters from approximately two dozen journals and three major book series over the past century. The project included myself, philosopher of science Evelyn Brister, and philosopher of science Rick Morris. We were initially interested in using this corpus to answer questions about gender and philosophy of science: How has the gender composition of the field changed over time? How does the gender composition vary across subfields of philosophy of science? Where and when have women contributed “germinal papers,” associated with the emergence of new subfields? While this project produced one contributed book chapter, “Contributions of Women to Philosophy of Science: A Bibliometric Analysis” (2020, with Evelyn Brister), it necessarily relied on name-based automatic gender attribution tools. Because of increasing concerns about the use of such tools, and the increasing prominence of trans and nonbinary philosophers of science (including myself), we judged that the potential benefits of this project did not outweigh the likely harms.

Version from 27 June 2019

Thanks to a series of academic and policy opportunities, I’ve been able to work on a variety of multidisciplinary research topics, using a variety of qualitative and quantitative methods. This page briefly describes some of my major research interests, including links to relevant publications. A full list of papers and other research outputs can be found on my ORCID profile.

Academic institutional effectiveness

Conceptual frameworks from philosophy of science, science and technology studies, and science policy; analytic methods from statistics and data science; and institutional data sources can be combined to support understanding of how science operates. Some of the specific questions I’ve examined in this area include:

  • Is novel toxicological research at the Environmental Protection Agency integrated into the broader research community?
  • How have interdisciplinary funding programs fostered novel collaborations in fields such as robotics and genomics?
  • What factors predict whether underrepresented groups will major in philosophy?
  • What conceptual frameworks can be used to understand the academic politics of research metrics?
  • How can article metadata and text mining methods be used to uncover forgotten contributions of women to philosophy of science?

Selected publications

Public scientific controversies

Public scientific controversies are my entry point into discussions of science policy and the role of expertise in democracy. Traditional models of such controversies focus on gaps between “scientists” and “the public,” and explain these gaps in terms of public ignorance and irrationality. In contrast, my approach to these controversies are based on ideas of power, structural oppression, and democratic accountability, in line with my background in feminist philosophy of science and political philosophy.

I am interested in many different specific issues that fall under the heading of “public scientific controversies.” Some of my current work in this area addresses ideals of transparency and fairness in algorithmic injustice, as well as controversies over obesity and pesticide regulation. My publications in this area are often written for policy audiences and published in venues beyond academic philosophy.

Selected publications

Science and values

This is the most disciplinary or narrowly philosophical topic among my major research interests. My dissertation (Notre Dame 2012) developed Alasdair MacIntyre’s conception of a social practice, informed by feminist philosophy, and applied this conception to the then-current debate over the ideal of value-free science. Since finishing my dissertation, I’ve become known for defending a distinctively Aristotelean version of the “aims approach” to values in science.

Due to the nature of my employment over the last several years, I’ve been unable to devote much time to narrowly philosophical work. I have been able to give presentations in which I apply and defend my version of the “aims approach” to topics such as academic freedom and “dangerous ideas,” open science, and objectivity. I intend to write up and publish these talks over the next few years.

Footnotes

  1. For comparison, based on a 2020 post at the philosophy blog The Philosophers’ Cocoon, at research-intensive universities assistant professors in philosophy are typically expected to publish 1-2 peer reviewed articles per year over a 5-6 year period (https://philosopherscocoon.typepad.com/blog/2020/01/what-are-your-universitys-tenure-standards.html).↩︎

  2. In retrospect, I should have approached Natalia about co-authoring the paper.↩︎

  3. I want to stress that citation-based metrics are problematic for all academic fields, but especially the humanities. Google Scholar has notorious data quality problems, and more reliable indices such as Clarivate’s Web of Science have extremely patchy coverage of humanities journals and, especially, books.↩︎

  4. As a biographical aside, I fondly remember writing the first draft of “Inductive Risk and Regulatory Toxicology” in the Jefferson Reading Room at the Library of Congress.↩︎

  5. As I explain in the published version, the paper was rejected because one reviewer and the managing editor insisted that there was obviously a replication crisis in environmental epidemiology, even after I showed that every citation they offered failed to provide such evidence.↩︎

  6. Specialists in the field typically prefer the more general term “scientometrics,” but “bibliometrics” seems to be more broadly understood among academics.↩︎