In addition to physics and astronomy, I used to study philosophy of science and sociology. In my opinion, many scientists could learn a few things from sociologists and philosophers of science, to help them to better understand and consider how scientific processes work, what influences them and potentially biases scientific results, and how science advances through their and others’ work. In addition, I think that people who aren’t professional scientists (who we often simply call “the public”) could better understand what we are learning and gaining from science and how scientific results are obtained. I’ll just write a few ideas here and we can discuss these issues further later, but my main point is this: science is an excellent tool that sometimes produces important results and helps us learn about the universe, our planet, and ourselves, but it can be a messy and nonlinear process, and scientists are human–they sometimes make mistakes and may be stubborn about abandoning a falsified theory or interpretation. The cleanly and clearly described scientific results in textbooks and newspaper articles are misleading in a way, as they sometimes make us forget the long, arduous, and contentious process through which those results were achieved. To quote from Carl Sagan (in Cosmos), who inspired the subtitle of this blog (the “pale blue dot” reference),
[Science] is not perfect. It can be misused. It is only a tool. But it is by far the best tool we have, self-correcting, ongoing, applicable to everything. It has two rules. First: there are no sacred truths; all assumptions must be critically examined; arguments from authority are worthless. Second: whatever is inconsistent with the facts must be discarded or revised.
As you may know, the title of this post refers to Thomas Kuhn (in his book, The Structure of Scientific Revolutions). “Normal science” (the way science is usually done) proceeds gradually and is based on paradigms, which are collections of diverse elements that tell scientists what experiments to perform, which observations to make, how to modify their theories, how to make choices between competing theories and hypotheses, etc. We need a paradigm to demarcate what is science and to distinguish it from pseudo-science. Scientific revolutions are paradigm shifts, which are relatively sudden and unstructured events, and which often occur because of a crisis brought about by the accumulation of anomalies under the prevailing paradigm. Moreover, they usually cannot be decided by rational debate; paradigm acceptance via revolution is essentially a sociological phenomenon and is a matter of persuasion and conversion (according to Kuhn). In any case, it’s true that some scientific debates, especially involving rival paradigms, are less than civil and rational and can look something like this:
I’d like to make the point that, at conferences and in grant proposals, scientists (including me) pretend that we are developing research that is not only cutting edge but is also groundbreaking and Earth-shattering; some go so far as to claim that they are producing revolutionary (or paradigm-shifting) research. Nonetheless, scientific revolutions are actually extremely rare. Science usually advances at a very gradual pace and with many ups and downs. (There are other reasons to act like our science is revolutionary, however, since this helps to gain media attention and perform outreach in the public, and it helps policy-makers to justify investments in basic research in science.) When a scientist or group of scientists does obtain a critically important result, it is usually the case that others have already produced similar results, though perhaps with less precision. Credit often goes to a single person who packaged and advertised their results well. For example, many scientists are behind the “Higgs boson” discovery, and though American scientists received the Nobel Prize for detecting anisotropies in the cosmic microwave background with the COBE satellite, Soviets actually made an earlier detection with the RELIKT-1 experiment.
Let’s briefly focus on the example of quantum mechanics, in which there were intense debates intense debates in the 1920s about (what appeared to be) “observationally equivalent” interpretations, which in a nutshell were either probabilistic or deterministic and realist ones. My favorite professor at Notre Dame, James T. Cushing, wrote a provocative book on the subject with the subtitle, “Historical Contingency and the Copenhagen Hegemony“. The debates occurred between Neils Bohr’s camp (with Heisenberg, Pauli, and others, who were primarily based in Copenhagen and Göttingen) and Albert Einstein’s camp (with Schrödinger and de Broglie). Bohr’s younger followers were trying to make bold claims about QM and to make names for themselves, and one could argue that they misconstrued Einstein’s views. Einstein had essentially lost by the 1930s, in which the nail in the coffin was von Neumann’s so-called impossibility proof of “hidden variables” theories–a proof that was shown to be false thirty years later. In any case, Cushing argues that in decisions about accepting or dismissing scientific theories, sometimes social conditions or historical coincidences can play a role. Mara Beller also wrote an interesting book about this (Quantum Dialogue: The Making of a Revolution), and she finds that in order to understand the consolidation of the Copenhagen interpretation, we need to account for the dynamics of the Bohr et al. vs. Einstein et al. struggle. (In addition to Cushing and Beller, another book by Arthur Fine, called The Shaky Game, is also a useful reference.) I should also point out that Bohr used the rhetoric of “inevitability” which implied that there was no plausible alternative to the Copenhagen paradigm. If you can convince people that your view is already being adopted by the establishment, then the battle has already been won.
More recently, we have had other scientific debates about rival paradigms, such as in astrophysics, the existence of dark matter (DM) versus modified Newtonian dynamics (MOND); DM is more widely accepted, though its nature–whether it is “cold” or “warm” and to what extent it is self-interacting–is still up for debate. Debates in biology, medicine, and economics, are often even more contentious, partly because they have policy implications and can conflict with religious views.
Other relevant issues include the “theory-ladenness of observation”, the argument that everything one observes is interpreted through a prior understanding (and assumption) of other theories and concepts, and the “underdetermination of theory by data.” The concept of underdetermination dates back to Pierre Duhem and W. V. Quine, and it refers to the argument that given a body of evidence, more than one theory may be consistent with it. A corollary is that when a theory is confronted with recalcitrant evidence, the theory is not falsified, but instead, it can be reconciled with the evidance by making suitable adjustments to its hypotheses and assumptions. It is nonetheless the case that some theories are clearly better than others. According to Larry Laudan, we should not overemphasize the role of sociological factors over logic and the scientific method.
In any case, all of this has practical implications for scientists as well as for science journalists and for people who popularize science. We should be careful to be aware of, examine, and test our implicit assumptions; we should examine and quantify all of our systematic uncertainties; and we should allow for plenty of investigation of alternative explanations and theories. In observations, we also should be careful about selection effects, incompleteness, and biases. Finally, we should remember that scientists are human and sometimes make mistakes. Scientists are trying to explore and gain knowledge about what’s really happening in the universe, but sometimes other interests (funding, employment, reputation, personalities, conflicts of interest, etc.) play important roles. We must watch out for herding effects and confirmation bias, where we converge and end up agreeing on the incorrect answer. (Historical examples include the optical or electromagnetic ether; the crystalline spheres of medieval astronomy; the humoral theory of medicine; ‘catastrophist’ geology; etc.) Paradigm shifts are rare, but when we do make such a shift, let’s be sure that what we’re transitioning to is actually our currently best paradigm.
[For more on philosophy of science, this anthology is a useful reference, and in particular, I recommend reading work by Imre Lakatos, Paul Feyerabend, Helen Longino, Nancy Cartwright, Bas van Fraassen, Mary Hesse, and David Bloor, who I didn’t have the space to write about here. In addition, others (Ian Hacking, Allan Franklin, Andrew Pickering, Peter Galison) have written about these issues in scientific observations and experimentation. For more on the sociology of science, this webpage seems to contain useful references.]