Dispatches from the Science Wars
I would like to begin with the story of my first encounter with the science wars. When I was an undergraduate studying physics at Manchester, my brother, who studied philosophy, got me interested in philosophy of science by asking me difficult questions over dinner.
Since the well-being of my psyche depended on being able to defend the position that physics is the most fundamental way of understanding the world, I jumped at the chance to take a course entitled “The Nature of Scientific Enquiry” when the opportunity came up. At the very least, I figured, it would help me win dinner table arguments with my brother.When the time came, I went to see my director of studies, the late Dr. Anthony Phillips (still the best physics teacher I have ever known) to tell him that I wanted to enrol in the course. His first reaction was, “Wouldn’t you rather take a course in fluid mechanics?”. After I rejected that option his next response was, “OK, but don’t believe a word they tell you.” At the time, I thought this rather uncollegial, but I did not realize that the course was run by the sociology department, and was being taught by a proponent of the “strong program” in the sociology of scientific knowledge (SSK), a major school on the sociology side of the science wars. I did not know that we were supposed to be at war, but, in light of that, Dr. Phillips comments make a lot more sense.
The first half of the course proceeded along the lines of a generic philosophy of science course. We studied Bacon [2], logical positivism [3], Popper [4], Kuhn [5], and Lakatos [6]. However, unlike a standard philosophy course, Kuhn was given a ringing endorsement, and then we went off to study SSK.
The strong program of SSK is most closely associated with David Bloor and his collaborators at the University of Edinburgh [7]. It is intended as a response to earlier approaches to the sociology of science, which are deemed “weak”.
In “weak” studies, sociological factors are only deemed important in understanding why “failed” or “false” theories are sometimes accepted. For example, one might look at how Stalin’s totalitarian rule allowed Lysenco’s ideas of environmentally acquired inheritance to become the dominant theory of genetics in the Soviet Union in the 1930s and 40s. In modern times, one might look at why the anti-vaccine movement or the idea that human activity is not causing climate change are being increasingly accepted in large segments of the US population.In contrast to this, the strong program states that sociological factors are equally important in understanding how successful scientific theories, which are usually deemed “true”, gain acceptance. If a theory is accepted science, it is very easy to fall back on the argument that the reason it became accepted is simply that it is “true”. Proponents of the strong program reject this asymmetry of explanation, and want to study the sociological reasons why science progresses the way it does period, without regard to whether a or not a theory is “true”. In order to do this they adopt, as a methodological principle, a ban on using the “truth” or “correctness” of a theory an explanation for its acceptance.
Although this ban is supposed to be merely methodological—a corrective for decades of studies which ignored sociological factors other than in cases of “error”— studies in the strong program tend to show strong sociological influences in every case they look at. Unless you are being deliberately contrarian, it is very hard not to infer that, if you can actually find sociological reasons why theories are accepted in every case, then scientific theories must be social constructs, with no claim to be the ultimate arbiters of objective truth. Although defenders of the strong program like to emphasize that the ban is meant to be methodological, and they are simply “hands- off” on the question of ultimate truth, it is pretty bizarre to adhere to a methodology and, at the same time, not contemplate the most obvious reason why that methodology might work well.
This leads to cultural relativism about scientific truth and, despite protests to the contrary, the language of cultural relativism does seep through the rhetoric of the strong program. Nonetheless, I define “sociologism” as the position that scientific theories are merely social constructs, in contrast to the strong program itself, which insists on only adopting this as a methodology. Sociologism implies that sociology is the most fundamental science, since it means that understanding the content of any scientific theory is equivalent to understanding the social factors that led to its acceptance.To see how easily SSK devolves into sociologism, I want to relate an experience from the Nature of Scientific Enquiry course. In one of our assignments, we were asked the question, “If sociological factors always play a role in determining which scientific theories are accepted, does science still tell us anything about the real world?” In the seminar discussion of this, the graduate TA proposed the answer, “Yes, because sociology is a science, so the study of sociological factors is still a study of the real world.” This is sociologism writ large. Not only do proponents of sociologism want to take physicists down a peg or two, but they also want to view their own subject as more fundamental than the sciences they are studying. Everything hangs off sociology, as it were.
It is easy to ridicule sociologism. After all, advocates of this view still get on airplanes to fly to conferences. If you really believe that science is just a social construct, then you have no good reason for believing the airplane will not simply fall out of the sky. I, for one, would not take the fact that flying airplanes is a tradition of my culture as a convincing argument to get on board. So, proponents of this view seem to act like they believe at least some aspects of science are objectively true, while simultaneously propounding the opposite.
In the throes of intellectual enquiry, it is common to adopt overly extreme views, which later have to be walked back.
This happens all the time on the speculative end of theoretical physics, e.g. the claim that the universe is literally a quantum computer [8], or that all entangled systems are literally wormholes [9], or that the universe is made of mathematics [10]. So let’s not hoist all of sociology on the petard of their most extreme proponents, and instead look at the evidence on which their claims are based.Most studies in the mould of the strong program proceed along the following lines. We first consider the modes of enquiry that are claimed to be the hallmarks of the scientific method, including such things as induction, falsifiability, the role of crucial experiments, skepticism of hypotheses that are not strongly supported by evidence, rational choice between programs of research, etc. Whichever of these (often conflicting) accounts of scientific enquiry you subscribe to, the sociologists find that they are violated in almost every case they look at, and identify sociological factors that played a role in theory choice instead.
There is not space to delve into specific examples here, so I will just mention Collins and Pinch’s study of the role of the Michaelson-Morely experiment in the acceptance of Einstein’s relativity [11], since that is of relevance to fundamental physics. In the usual story told to students, the Michaelson-Morely experiment is a crucial experiment that led physicists to reject the luminiferous ether, i.e. the idea that light waves must propagate in some medium in the same way that you cannot have water waves without there being some water to do the waving. The ether was replaced by Einstein’s theory, which eliminates it. Collins and Pinch show that Michaelson- Morely experiments never produced conclusive evidence against the ether, despite attempts spanning several years under different experimental conditions.
Now, one might argue that the weight of experimental evidence for relativity that has been acquired since then is justification for its acceptance today, but still it was accepted long before any of this was acquired.
One might also argue that Einstein’s theoretical explanation of the symmetry of Maxwell’s equations is the real reason why relativity was accepted, but this was not universally regarded as compelling at the time. Indeed, the controversy over this is the reason why Einstein won the Nobel prize for his explanation of the photoelectric effect rather than for relativity. While it is a stretch to conclude from this that relativity is just a social construct, the process of its acceptance was rather less rational than one might otherwise believe. At the very least, the story we tell about how relativity became accepted, which is part of the pedagogy of relativity, is largely a social construct.However, the problem with case studies like these is that philosophical theories of science are not supposed to have the same status as mathematical theories. In the latter, if you find one counter-example to a theorem then the theorem is false.[1] Instead, philosophical theories of science propose norms, which we should strive to adhere to if we want to create reliable scientific knowledge. These norms include skepticism of hypotheses that have no evidential support, designing experiments that remove as much bias as possible, etc. Nobody is claiming that these norms are strictly adhered to 100% of the time, and that sociological factors play absolutely no role. Instead, the claim is that by attempting to adhere to these norms, the community as a whole, over long periods of time, will develop knowledge that is more reflective of the objective world than otherwise.
To put it another way, the “scientific method” cannot really be characterized in a precise way that is applicable to all cases. For any methodological principle that you might propose, one can find cases where it is not really applicable. But that does not mean that, upon looking at the particulars of a specific theory, one cannot decide whether the evidence supports it. We may use different methods and standards of evidence in fundamental physics, climate science, and psychology, but these all bear a family resemblance, and an expert in one of those fields can use the available evidence to decide how likely a given claim is to be true.
The fact that we cannot give a discipline-invariant definition of the scientific method does not seem to have gotten in the way of the progress of any scientific discipline in particular.Nonetheless, the studies of the strong program do show that social factors have played a larger role in the construction of “true” theories than you might otherwise have thought, so the idea that we should only pay attention to sociology in cases of “error” is suspect. Generally, all scientific discourse takes place within a language, and is conducted by entities that are situated within a society, with all the baggage that entails, so social values are implicitly used in the construction of science whether we like it or not. Although physics makes heavy use of mathematics, so is arguably less influenced by the particulars of common language than other sciences, few physicists believe that the content of physics is entirely contained in its mathematical equations. We need discourse to understand what our theories mean, how they are connected to observations, and even what questions are sensible to ask of the them. Hence, the idea that physical theories may not be completely objective, and that sociological factors may play a role in their very construction, should at least be an option on the table, regardless of how small or large you think that role is.
One example where sociological factors have had a strong influence on physics is the dominance of the Copenhagen interpretation in the foundations of quantum mechanics. To modern eyes, it looks like the founders of quantum mechanics jumped to conclusions about the nature of (un)reality based on scant evidence. While much evidence that can be construed as supporting this kind of view has been acquired in the meantime, the Copenhagen view was accepted by the majority of physicists for decades without many physicists actually feeling the need acquire this evidence. Although there is more tolerance for diverse views on the interpretation of quantum mechanics today, Copenhagen has had a lasting influence on what physicists think a physical theory should look like, which may be cutting off fruitful research directions.
On the other hand, we do not want to endorse sociologism, in which we cannot explain why airplanes do not fall out of the sky, why children should be vaccinated, and why we should take action on climate change. The success of our fundamental physical theories surely means something for the objective physical world. Therefore, we should not replace the claim that physics is fundamental with the claim that sociology is fundamental instead. What we need is a theory of knowledge that can account for why we should trust that airplanes will not just fall out of the sky, but also allows external factors to influence physics in a controlled way. If it can also explain why smart people can be led to believe that physics is fundamental, and other smart people that sociology is fundamental, then so much the better.
3