Kevin Elliott’s A Tapestry of Values is a much welcome, accessible, case-based exploration of the questions that are central to the literature on values in science intended for a general undergraduate audience. Elliott writes that the book is meant to address two problems he saw in the current offerings on values in science: “First, it sometimes emphasizes the general point that science is deeply influenced by values but without getting into nitty-gritty details about the precise ways in which values influence specific choices about scientific methods, concepts, assumptions and questions. Second, it often highlights the presence of values in science without clearly exploring how to tell the difference between influences that are acceptable and those that are not” (ix). Both are pressing issues for the field, but the first is more thoroughly dealt with than the second in the current literature (as the author himself notes). Nitty-gritty details abound in Tapestry.

Elliott’s work has the virtue of including a clear description of values in his first chapter: “Broadly speaking, a value is something that is desirable or worthy of pursuit. So, for example, scientists typically value accurate predictions, clear explanations, logical consistency, fruitful research projects, honesty, credit for their accomplishments, health, economic growth, environmental sustainability and global security” (11). While this seems to be a largely serviceable definition, Elliott does not link his description of values to fact/value distinction works by figures like Hume and Putnam—one of the most common philosophical uses. In addition, this kind of description makes it difficult to see why anyone might object to talking about values in science, which has been an important issue in the field.

Tapestry is organized around six main chapters, with an additional introduction and conclusion. These six chapters each focus on a fundamental question: What should we study? How should we study it? What are we trying to accomplish? What if we are uncertain? How should we talk about it? How can we engage with these values?

After an introductory chapter that uses the traditional case study of Vavilov and Lysenko’s work in Soviet agriculture as well as Colborn’s more recent work in environmental science to introduce broad issues of values in science, chapter 2 focuses on the question “What should we study?” This chapter centers on questions about the prioritization of gender/racial cognitive difference research, here illustrated by Larry Summer’s resignation at Harvard following comments about innate differences between men’s and women’s aptitude for math and science. The chapter also takes on the question of public funding for science, focusing on recent attempts by Republican congressmen to increase political oversight over the National Science Foundation (NSF) and other agencies. While Elliott claims that publicly endorsed values should influence publicly funded science, he also suggests that members of Congress are not in a good position to be “double-checking grant proposals at the NSF,” given a lack of expertise and the temptation of shutting down particular programs for political gain. The chapter finishes by examining the case of malaria to critique the way values in private research result in the production of biomedical interventions that are largely focused on diseases that are common in wealthy nations to the detriment of low-income countries.

Chapter 3, entitled “How Should We Study It?” focuses on case studies involving golden rice, pharmaceutical research in depression, and environmental pollution in Boston, Massachusetts, and Madison, Wisconsin. In each case, researchers and others operating with very different methodologies, assumptions, and questions come to very different conclusions. For example, if you assume that the color of rice makes no difference to its rate of consumption, that industrial farm practices are the norm, and that people in poor countries have little access to expensive fresh produce (resulting in vitamin A deficiency), then “golden rice” can look like a great benefit to society. But if you think that the populations in question are unlikely to eat strange-colored food, that industrial farm practices resulted in devastation for traditional farmers whose methods were better matched to the land they worked with, and that people in poor countries have access to nontraditional sources of fresh produce (when it is not destroyed by industrial farm practices), then “golden rice” is an unwanted technological answer to a political problem.

Chapter 4, focuses on the question “What are we trying to accomplish?” highlighting that the aims of science may be diffuse and perhaps even in conflict. For example, The Rosen classification system (Natural Channel Design, or NCD) for river damage and restoration is simple and cheap, but fails to capture nuance that other systems prize. If our goal is to quickly and inexpensively assess damage to river ecosystems and make recommendations regarding restoration, says Elliott, it makes sense to use this system. If we are concerned that these assessments be as reliable and complete as possible, NCD is not a good classification system. This case study is meant to problematize the assumption that what science is consistently up to is “working with the most reliable, accurate theories or models that they can develop” and claiming instead that different projects have different aims. The NCD story is paired with additional case studies on “wetland banking,” the choice of a discount rate in the Stern report on the economic impact of climate change, and trade-offs between general circulation models and regional climate models in climate change modeling. This chapter’s short section on “some objections” is tellingly philosophically rich and yet unlikely to satisfy entrenched critics. Here, Elliott attempts to defuse worries about the problem of wishful thinking by arguing that scientists can legitimately use values in order to decide which theories to pursue, while still holding that it would be inappropriate for scientists to use values to decide which ideas to accept. This hangs on the reader accepting that there might be reasons for pursuit that do not hang on possible future acceptance. Without this strong difference between pursuit and acceptance, Elliott’s insistence that, while it might be inappropriate for values to direct acceptance, it is acceptable for values to motivate pursuit cannot get off the ground. While there might be resources in the history of philosophy of science to support such an argument, they are not presented here, perhaps for considerations of brevity. Given that Elliott has argued elsewhere that values can legitimately influence what ideas to accept (e.g., Kevin Elliott and David Willmes, “Cognitive Attitudes and Values in Science,” Philosophy of Science 80, no. 5 [2013]: 807–17), this is a surprising direction for this author.

In chapter 5, “What If We Are Uncertain?” Elliott argues both that values are relevant to respond to uncertainty in science and that in some cases values can increase uncertainty in science. This chapter features case studies on Hansen’s 1988 testimony on climate change to congress, Colborn’s work on pollution as a cause of endocrine disruption and therefore rising breast cancer rates, and Douglas’s philosophical explanation of values in dioxin studies. All three can be taken as instances of trade-off between objectivity and social values. Understanding this tension-filled trade-off requires first accepting that objectivity is a value itself, rather than the virtuous absence of values, which may not be clear to Elliott’s intended general audience. In addition, Elliott categorizes what is happening when people resist accepting that smoking causes cancer or believe that vaccines cause autism as an example of values “generating or contributing to” uncertainty in science. While I can see how it might be tempting to suggest that this phenomenon is the mirror image of those discussed earlier in the chapter, it seems to me that entrenchment phenomena are not cases in which values are adding to uncertainty in science but a case in which values are blocking acceptance of a theory. This modification would fit with his claims elsewhere in the book and would give him additional resources to criticize so-called merchants of doubt. Indeed, throughout the book Elliott often takes care to bring to light the values in play but often does not include any determination about the positive or negative effects of these values. This may strike some readers as admirably even-keeled and others as a strange unwillingness to take a stand on values that were and are destructive.

Chapter 6, “How Should We Talk About It?” includes an example about the ways research on monogamy in voles has been assumed to generalize to tell us interesting things about infidelity in humans. This example allows Elliott to discuss the ways in which frames, terms and metaphors, and categories used in scientific communication involve values. Elliott writes, “They [vole researchers and writers covering this research] are giving the impression that there is a fairly straightforward relationship between particular genes and people’s behavior, but claims like these are premature” (114). Elliott goes on to argue that it is best not to think of this as simply sloppy or inappropriately value-laden science or science reporting. Instead, he suggests there is a fundamental tension here, a tension that recalls arguments from chapter 5: “On one hand, if scientists simply ‘stick to the data’ and refuse to reflect on the significance of their findings, they risk depriving the public of valuable information. On the other hand, if they deliberately try to frame their findings in ways that connect them to broader public concerns, they risk coloring the research with their personal values and worldviews” (115–16). The chapter includes practical strategies for improving science communication—for example, “backtracking,” that is, framing problems in such a way that it is possible to recognize the frame in use, and acknowledging problems with/alternatives to this frame. An additional highlight of this chapter is the discussion of the work of Mildred Cho and others to suggest that racial categories do not usefully map on to important disease-tracking genetic variations and Jay Cohn’s work defending the use of racial categories when tackling sickle cell disease and health inequalities.

Community engagement with values in science is the subject of chapter 7, “How Can We Engage with These Values?” This chapter includes case studies that focus on soliciting and interpreting values from different stakeholders in different scientific projects, from the efforts of ACT UP activists to modify the testing and development of drugs to treat HIV and AIDS to public input into nanotechnology research. Four basic types of engagement are discussed: bottom-up public engagement, top-down public engagement, interdisciplinary engagement among scholars, and engagement with institutions.

In a final concluding chapter, Elliott highlights the earliest grassroots environmentalism of Commoner, explains his tapestry metaphor, and begins to articulate a set of criteria for the appropriate use of values in science. These criteria are engagement, transparency, and representativeness. This last chapter could helpfully direct students to next texts on the management of values in science and can provide some satisfaction to readers who quickly grasp the kind of empirical claim that Elliott makes (there are values in science) and are more interested in moving on to a normative claim (here is how we should deal with values in science).

While the book hangs together well as a whole and its brevity is generally a selling point, it is tempting to point out what might have been elaborated on. It might have been interesting for Elliott to have more explicitly connected these claims about values in science to their implications in epistemology. For example, Catherine Elgin’s True Enough (Cambridge, MA: MIT Press, 2017) seems to be a natural fit, given her interest in accounting for epistemic practices in science that succeed while not being overly concerned with verity.Footnote ¹ However, it is understandable that the author did not pursue this connection, given the goal of producing an accessible case-based introductory volume.

The nature of values themselves as intentional or unintentional, knowable or unknowable, is only briefly discussed. This leaves the reader unable to answer some very important questions: What kind of methods should be used for determining what values are in play in a scientific project? Should we be polling scientists for their self-identified value commitments? Should we use some kind of behavioral model, supposing that some actions indicate a particular underlying value commitment?

It is no secret that academic book reviews, like this one, which call their subject “accessible,” often do not mean this as a sincere compliment. This book is accessible. This is a sincere compliment. It would be an excellent fit for a wide range of courses (Science and Society, Introduction to Philosophy of Science, Science Policy, Research Ethics in various sciences) in a wide range of disciplines (philosophy, sociology, science and technology studies, liberal studies, environmental studies) and is an important addition to scholarship on values in science.

While other reviewers of this work (Erin Nash, “Are Values in Science Like a Tapestry or a Patchwork Quilt?” Science and Education 26, nos. 7–9 [2017]: 1063–69; Heather Douglas, “From Tapestry to Loom: Broadening the Perspective on Values in Science,” Philosophy, Theory and Practice in Biology 10, no. 8 [2018]) have substituted their own textile-based metaphor for the role of values in science, I think Elliott gets it mostly right with his tapestry explanation: “the tapestry metaphor highlights at least three important lessons about scientific practice: (1) the analytical or rule-governed components of scientific reasoning are deeply interwoven with components that are influenced by values; (2) the roles of values can be analytically disentangled from other aspects of scientific reasoning; and (3) specific influences of values can have ‘ripple’ effects across science that require careful attention.”Footnote ² A book that discusses all three, while remaining case oriented, is very welcome in our field.