Philosophical works about scientific realism often point to public trust in science as a key motivating consideration. The intuitive idea is that if we have reason to think scientific theories are approximately true, then we have all the more reason to think our trust in science is warranted. Yet this raises the concern that philosophical challenges to realism may further damage trust in science. Should we worry that arguments for antirealism will engender mistrust in scientific findings about climate change, vaccines, or evolution? Additionally, how can philosophers who are antirealists consistently maintain a pro-science attitude?
Debates around scientific realism explore questions such as whether theoretical terms refer to things “out there” in the world; whether we’re justified in believing that scientific theories are (approximately) true; and the extent to which scientific concepts are shaped by social forces and historical contingencies rather than objective reality. I argue that questions about scientific realism actually have no relationship to whether science-based recommendations are trustworthy. I suggest that this is a good thing for trust in science.
I’ve been thinking about an underlying tension in the work on scientific realism. Most philosophers of science are naturalists. Roughly, most of us endorse the claims that science is our best source of knowledge about the world, that scientific explanations of natural phenomena are superior to supernatural explanations, and that philosophical views should be developed in meaningful conversation with scientific findings. However, views about scientific realism contain metaphysical claims which extend far beyond the reach of contemporary scientific practices.
Realist and antirealist positions include claims about metaphysics (e.g., whether underlying reality is material, structural, or ideal); about the extent to which objects are “given” or constructed; and about the possibility of terms in our language tracking real objects despite significant conceptual change over time. I call these “deep” metaphysical issues. These issues have to do with the underlying nature of reality and our ability to epistemically access and represent that underlying reality. Yet, scientific practices cannot help us to answer these questions.
Let me provide a brief thought experiment to make this claim plausible. Think about running an experiment in a chemistry lab. Two substances are mixed, causing a reaction which is presumed to release some gas. You take some measurements in order to calculate how much gas has been released. Your measurements correspond to what’s predicted by theory, allowing for a reasonable amount of error.
Here is the key part of the thought experiment: suppose that Berkeley was right and everything in the world is ideal. By hypothesis, contrary to what most people assume, your chemistry experiment did not involve any material substances. Instead, you’ve been mixing and measuring ideas. But your experiment functions exactly the same as it would if the stuff were material. You make the same observations, measurements and calculations and draw the same conclusion. This means that the truth or falsity of Berkeleyan idealism can have no detectable influence on scientific practices.
So, I make two related claims: that scientific practices and deep metaphysics are insensitive to and lack evidentiary relevance to one another. By insensitivity I mean that scientific practices could function in exactly the same manner regardless of the underlying metaphysics (e.g., if the world were fully ideal, fully material, or constructed in a Kantian way). By evidentiary irrelevance, I mean that scientific practices cannot supply evidence for or against these deep metaphysical claims. Evidentiary irrelevance is a consequence of insensitivity. Compare: if a thermometer’s reading is insensitive to the size of the human population, then thermometer readings don’t have evidentiary relevance when estimating the human population.
If this is right, the troublesome consequence is that it’s impossible to justify views on the scientific realism spectrum by appealing to any features of science. Science cannot supply any evidence for or against these views. This results in the tension I mentioned before: we can’t supply evidence for or against scientific realism which meets our own standards for good evidence about the natural world.
Although some data show a majority of philosophers lean towards scientific realism, antirealist positions (like Cartwright’s, Kuhn’s, and van Fraassen’s) have exerted a major influence over the discussion, and the literature shows no signs of the issue being resolved to everyone’s satisfaction. I think a major factor contributing to the intractability is that philosophers of science are trained to examine philosophical ideas in the light of scientific practices, yet per my argument scientific practices shed no light on this big problem. What should we conclude from all this? My own conclusion is that we should adopt quietism—the view that we cannot decide which view is right. But I also want to return to the issue about trust in science.
The thought experiment above shows that science can function exactly as it does regardless of the underlying metaphysics. You can apply the same thought experiment not only to idealism but also to instrumentalism, constructivism, pragmatist views, and different theories about the nature of truth and reference (i.e., whether and how scientific terms grab on to external reality). You can even apply the thought experiment to favorite skeptical scenarios, like brain-in-a-vat or the hypothesis that we’re in a simulation. If none of these theories is inconsistent with current scientific practices, then the success of science is insensitive to all these metaphysical possibilities. If social and epistemic mechanisms producing objectivity, convergence, and replicability generate trustworthy results independent of the underlying metaphysics, our trust in science should not depend on issues of scientific realism or even metaphysical realism. This is good because philosophical threats to scientific realism therefore can’t undermine our trust in science. Scientific practices work exactly as well as they do regardless of what the underlying world is like.
Katie Morrow
Katie Morrow is a postdoctoral researcher in the philosophy department at Bielefeld University, where she is part of an interdisciplinary DFG-funded Collaborative Research Center studying the ecological niche. She received a PhD in History and Philosophy of Science from the University of Pittsburgh in 2022.
