In this post, I discuss the role of scientific explanation in explaining the historical development of science. I argue that we often need scientific knowledge to provide such explanations and therefore the relationship between historical explanations and future estimations is complicated in the case of science.
(More detailed analysis of the issues in this post is presented in my 2019, Section 6.8. https://www.utupub.fi/handle/10024/147334)
In the previous post, I discussed the “pull of the nature” intuition. The intuition says that because world the world has a definite structure that affects our experiments and observations, sooner or later that reality shapes our theories to be adequate descriptions of it. We noticed that, while this intuition is problematic, it seems to capture something important about science. However, there exists an opposite position. In the “strong programme” in the sociology of science, it has been claimed that the way the world is – the nature – cannot explain our beliefs about it.
David Bloor writes:
If we believe, as most of us believe, that Millikan basically got it right, it will follow that we also believe that electrons, as part of the world Millikan described, did play a causal role in making him believe in, and talk about, electrons. But then we have to remember that (on such a scenario) electrons will also have played their part in making sure that Millikan’s contemporary and opponent, Felix Ehrenhaft, didn’t believe in electrons. Once we realize this, then there is a sense in which the electron ‘itself’ drops out of the story because it is a common factor behind two different responses, and it is the cause of the difference that interests us. (Bloor 1999, 93).
Let’s use an existing account of explanation to analyze what goes wrong in this argument.
In a counterfactual account of explanation (I rely on Woodward 2003) “the underlying or unifying idea in the notion of causal explanation is the idea that an explanation must answer what-if-things-had-been-different questions, or exhibit information about a pattern of dependency” (2003, 201). To put it simply, explanations answer questions of the form “Why X rather than Y?” by pointing out factors Z and W such that “had W rather than Z been the case, Y rather than X would have been the case”. For example, we can explain why I wear a black shirt (X) rather than red one (Y) by noting that the red one was wet (Z): Had the red shirt not been wet (W), I would have worn it.
In the light of counterfactual account of testing, it is not true that electrons do not explain Millikan’s belief. Had there not been (free) electrons in Millikan’s device, he would not have had the belief. (The local absence of (free) electrons could have been brought by omission of ionizing radiation, ionizing radiation being an important part of the experimental setup). Of course, the belief in electrons cannot be explained by electrons alone – otherwise everyone who ever lived would have believed that such things exist. However, this does not mean that the electrons are not a part of a satisfactory explanation for the beliefs. We do not think that throwing a match is not a cause of a fire because gasoline was also involved. Moreover, if there were no electrons, neither Millikan nor Ehrenhaft would have believed in them. This shows how electrons explain the difference in their beliefs: had there not been electrons, there would have been no difference in the beliefs.
These technical issues take us to a more important discussion. Scholars have though that there is an epistemological concern: Since we believe in electrons because of Millikan’s work, we cannot use our belief in electrons in the explanation of Millikan’s belief without circularity. We have no access to the reality that could confirm – independently of our Millikan-based tradition – whether or not electrons exist. (E.g. Kochan 2010, 136.)
In my view, there is no circularity here. It is not our belief about the electron that the explanation cites but the electrons. If electrons exist, then Millikan was exposed to them and our electrons-citing explanation of his belief is right. Whatever our level of confidence toward the existence of electrons is, the truth value of the explanation does not change. Most certainly we cannot say that electrons did not affect Millikan because we are not absolutely sure that they did. Ignoring a possible factor is never the right way to proceed.
It also seems to be the case that we cannot justify a theory about some entity without at the same time committing ourselves to the explanatory status of that entity. We ask “What would an entity E cause in experimental setup S if E existed?”. If we then infer that it would cause phenomenon P, and if we observe P, then it seems that we have evidence for the explanatory status of that entity. (The exact form of the inference need not bother us here; the point is that it is difficult to see how there could be evidence of the existence of an entity without some (supposed) causal link between the entity and the evidence). If, after many such inferences, we accept the theory about the entity, then we, at the same time, accept that it was a causal factor in our setup. It would not be reasonable to ask “Well, we accepted the theory about the entity. But does it have any causal role to play?” as if we could separate these issues.
This is important for the prospects of estimating the possible futures of science. If it is indeed the case that a (satisfactory) historiographical explanation of why some E was discovered requires that we cite how E causally interacted with instruments and researchers, then such historiographical explanation cannot be given without citing a scientific explanation of the working of E. The problem is not that we cannot explain the discovery before it was made since no historical event can be explained before it happens. Rather, the problem is that no matter how many historiographical explanations of different discoveries we have, a new case always requires information that is not incorporated in the previous explanations. In many cases we can reasonably infer that because Z explained X before, and we are now in the situation Z, X will be the case. For example, if we know that I used a black shirt because the red one was wet, then we can infer that if the red shirt is wet tomorrow, I will use the black one. Such inferences are of course always fallible, no matter how well justified, but they are nevertheless reasonable. However, in the case of scientific discovery, it seems that we can never know whether we are in the situation Z (interacting with a phenomenon P with our instruments) before we have made the discovery X (that the phenomenon has the properties Q). Again, in order to know that we are interacting with a phenomenon in a way that leads to a discovery, the discovery should have been already made which is impossible. We can explain Millinkan’s belief by citing the properties of electrons and how they affect the instrument. However, Millikan could not have said that “because I am now interacting with electrons and their properties Q, I am about to make the discovery that they have the properties Q”. He should have known what he was about to discover before the discovery was made.
Or perhaps Millikan could have legitimately said that. After all, hypotheses drive scientific research and discoveries. For example, Millikan made the hypothesis that the charge is not continuous variable and then discovered that this indeed is not the case. It seems that we can use potential scientific explanations to estimate the futures of science. However, our estimates are no more justified than the scientific hypotheses before they are tested. Again, it seems that estimations of futures of science and scientific hypotheses coincide.
The final issue that I want to mention is that one could argue that we cannot estimate the future developments in any area of life on the basis of the explanations of the past developments since every historical moment is unique in the same way as every scientific discovery is unique: it includes something that we could not have known before living through that moment. I am skeptical about this “historical uniqueness”. It seems that there repeating patterns in many areas of life and it is difficult to understand how we could make sense of anything if every moment was unique. Moreover, there have been genuine explanations in many areas of life (for example, why Mike Tyson lost to James “Buster” Douglas) that can be projected in the future (if your cornermen do not bring an endswell or ice packs, you might be in trouble). In general, it seems that we can estimate many areas of life on the basis of scientific explanations. We simply use the same explanatory information again and again. However, when it comes to the future discoveries of science, one cannot use scientific explanations to casually estimate them because the discovery may change the stock of explanatory information we have. That is seems to be the difference between science and other areas of life when it comes to futures: In other areas of life, it is possible, at least in principle, to use scientific knowledge in estimating the future, whereas the future of science concerns the creation of that knowledge. One cannot eat a cake that one has not baked.
Bloor, David (1999). “Anti-Latour”. Studies in History and Philosophy of Science 30 (1).
Kochan, Jeff (2010). “Contrastive Explanation and the ‘Strong Programme’ in the Sociology of Scientific Knowledge”. Social Studies of Science 40 (1).
Woodward, James (2003). Making Things Happen. A Theory of Causal Explanations. Oxford University Press.