File Under: Philosophy of Science
1. Introduction
We raise our children to think of science as test tubes and white lab coats and once they’re a bit older, we broaden that definition by teaching them about the scientific method. “Science”, we tell them, “is about making observations and forming hypotheses about those observations”. Of course, that is too simplistic a description of science, which is indicative of not only the rich history of scientific study but also the breadth of its subject matter. Defining what science is, how it is executed, by whom and in what context is an incredibly specific endeavor taken on by philosophers of science for hundreds of years, and those issues are doubtful to ever be solved. This paper only seeks to define a specific instance of, and dividing lines for, a particular circumstance within the realm of scientific study that is subject to much obfuscation: Can scientists posit the existence of an unobservable entity? Until the invention of the microscope, germs were unobservable but very real. Much was theorized about the existence and behavior of atoms, but until the development of the electron microscope, atoms had never been directly observed. But for a person of science to believe in the existence of something that is unobservable requires some level of faith in a theory, and that seems to fly in the face of what we think of as traditional scientific practice. This issue is called by many names but throughout this paper will be referred to as the Theory-Observation distinction. After a brief disclaimer, I will establish a contextual when for these varieties of thoughts, as well as who for the major players.
1.1 Disclaimer
On the issue of the theory-observation distinction, there are swaths of scholarly articles and books written by deep thinkers from just the last sixty years or so. I managed to read only a handful of them and feel as though I finally learned enough to know that I am completely unqualified to speak with any authority on the subject. To me, the structure and tone of “research paper” carries with it an air of sophistication and authority that I find difficult to balance against the motley collection of information I have to share. One of the questions left unanswered for me is whether this argument is one of semantics; is this a descriptive or a proscriptive argument? 1. Are we arguing about what science does for us; how we can refer to, and what can we infer by, its findings? Or are we attempting to separate the wheat from the chaff- to demarcate science from “other”? (Longino, 2006). I have some proofs of both possibilities and I will attempt to present them sensically to you, reader, if you will accept this disclaimer. Now back to our regularly scheduled programme.
1.2 The Major Players
The question of studying the unobservable has powerful implications: how is a line drawn between science and pseudoscience, and between theoretical claims and blind faith when much of a particular scientific process is intangible? This question is the subject of disagreement in the philosophical community and gave birth to a new set of ideas from philosopher Bas van Fraassen which became collectively known as Constructive Empiricism. These thoughts are a response to the Scientific Realist position that has evolved from the Logical Empiricist movement.
2. Background
2.1 Logical Empiricism
One can appreciate the desire of Logical Empiricism to limit itself to strict observation statements and theory based on deductive reasoning. Afterall, isn’t science meant to describe the world around us and define the mechanisms by which the world works? And considering the relative simplicity of the sciences in the 19th century and prior, this approach is unquestionably pragmatic. However, as the sciences began looking at the very small or very large, unobservable entities began to present a significant problem to Logical Empiricists. As scientific tools for measurement, mathematics, and the like became increasingly complex, and suggested the existence of something that was not yet detectable, it became necessary to evolve a scientific approach that could account for observationally challenging subject matter.
2.2 Scientific Realism
What better name to adopt for one’s progressive approach than realism? Philosopher Nora Mills Boyd deftly summarizes this pivot in philosophical approaches: “[Logical empiricism] was characterized by the aim of conceptually separating theory and observation, so that observation could serve as the pure basis of theory appraisal. More recently… we see philosophers… embracing a view of science in which the theoretical and empirical are usefully intertwined.” (Boyd, 2009). When experiments were first being conducted on the existence and behavior of electrons in 1909, repeatable and reliable data was collected on the behavior and properties of a thing that was (and continues to be) too small to observe. Presumably, a logical empiricist would reject the theory of the existence of an electron because the phenomenon only pointed toward a something, and that something needed to be seen, touched, or observed in some other way before accepting that the theory to be true. Clearly, logical empiricism had run its course and there would need to be more accommodating standards by which science could continue with its work.
2.3 Constructive Empiricism
In 1980, Danish philosopher Bas van Fraassen published The Scientific Image. The purpose he specifically states for the text was to develop a constructive alternative to scientific realism. His account of realism was that “theories account for the phenomena by postulating other processes and structures not directly accessible to observation…” and his primary objection is that there is an implication that this “science aims to find a true description of unobservable processes that explain the observable ones” (Fraassen, 1980). The sticking point for Fraassen is around the truth of the theory, as opposed to the usefulness of the theory. Acceptance of a theory from a realist perspective is to accept its truthfulness, while acceptance of a theory from a constructive perspective is to accept its utility. This where the discussion appears to settle into the epistemically semantic, but we will continue for now. Fraassen offers two very contrastable statements of intent from the two camps. First realism: “Science aims to give us, in its theories, a literally true story of what the world is like; and acceptance of a scientific theory involves the belief that it is true.” and for constructive empiricism “Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate” (Fraassen, 1980, p. 8,12).
2.4 The Sticking Point
We may intuitively know the difference between observable and unobservable, but it is still important to clearly define before moving forward, and Chakravarty has a way with words: “The distinction between the observable and the unobservable reflects human sensory capabilities: the observable is that which can, under favorable conditions, be perceived using the unaided senses (for example, planets and platypuses); the unobservable is that which cannot be detected this way (for example, proteins and protons).” (2011). However, there is a far more important distinction between unobservable and unobserved. The unobserved is another way of saying not-yet-observed, but the importance of this distinction to the empiricist is that a theory based on an unobserved entity will be either confirmed or denied once the thing is observed (Maxwell, 1962). Take a blind date, for instance; with only a photo and brief biography in a dating app, you imagine how the voice of this person might sound once you meet them, or how they may smell, or what their hands feel like in yours. You’ve constructed a theory that 1) this person exists, even though you have very little evidence that they do 2) and that your senses will detect certain things about this person: particular pitch of voice, the fresh smell of laundry on their clothing, and warm soft hands. No doubt, you’ve constructed the sensory data in your head because it is what you hope to, or have faith that you will, experience in the encounter. These are all allowable within the framework of the empiricist because you will meet the person and your theories will either be confirmed or falsified. In an unobservable case, you would not have a dating profile with which to create an expectation of your perfect match. Instead, you would hold a theory that there is a soul-mate, a perfect match, somewhere out there and the only thing that could confirm that theory is by meeting (observing) that which very likely does not exist. Additionally, to overturn that theory, one must observe (meet) every possible candidate for that soul-mate, which is obviously impossible to do.
3. Arguments
3.1 Fraassen’s Issue with Realism
As previously discussed, Fraassen’s primary concern with the realist account of theory is that of the implication of truth. Comparing the views of anti-realism, he makes the distinction that “according to the realist, when someone proposes a theory, he is asserting it to be true. But according to the anti-realist the proposer does not assert the theory; he displays it, and claims certain virtues for it. These virtues may fall short of truth: empirical adequacy, perhaps; comprehensiveness, acceptability for various purposes.” (Fraassen, 1980, p. 10) When it comes to proposing a theory based on an unobservable entity, Fraassen favors the logical empiricist position of True statements being reliably True, so that there is no room for interpretation of the defining language. He uses an example, of which I personally find quite telling, of his discomfort around the ambiguity of theory language in unobservable contexts, the Christian bible. While Fraassen’s critics are quick to point out all the instances that unobservables have retroactively become defined as unobserved in the history of science, he points to the scenario where literal interpretations of the bible by fundamentalist Christians lead to problematic conclusions such as the age of the earth. To be fair, he is not saying that science and theology are the same and should be treated the same. In this passage he is simply trying to define what he means by literal. However, I find that invoking theology to define the meaning of a word in a completely separate context is a bit of a linguistic trick. Moving on from the presentation of the theory to the arrival to the theory, Fraassen argues that without the sufficient empirical support for a theoretical entity, one can only arrive at the theory through what is most likely to be true, which is referred to as inference to the best explanation, and that “inference to the best explanation is an unwarranted form of inference and, therefore, belief in the existence of unobservable entities is always unwarranted” and “belief in the existence of unobservable entities ultimately relies on an unwarranted inference to the truth of the theory that postulates their existence and, therefore, is not adequately supported.” (Contessa, 2006). I think that is very well put by Gabriele Contessa and provides a nice segue to arguments against Fraassen’s constructive empiricism.
3.2 Arguments against Fraassen
Contessa counters Fraassen by suggesting that perhaps there should be separate categories for allowable language in theory postulation. She suggests that there should be “three main kinds of ontological commitment: ontological commitment to an individual entity, ontological commitment to a natural kind, and ontological commitment to a theoretical kind” (Contessa, 2006). The natural kind and theoretical kind are the objects of our focus, and by her own admission are difficult to differentiate. I prefer to skip straight to the definition of a theoretical kind as our first bucket and relegate everything else into the second bucket that we need not define. She says, “In the case of dark matter, the theory postulates the existence of a kind that does not correspond to any natural kind to which we are already committed.” (Contessa, 2006) which I understand to mean that the theory of dark matter is not redefining what has already been defined in some other way, but instead in postulating the theory of a new kind. If the theory of dark matter claimed that electrons account for the “missing matter” in the universe, and that is made possible by an error in the calculation of the mass of an electron, commitment to that theory would be unwarranted because it is opposed to what we accept to be true of a thing that exists (though it is not-yet-observed). In the case of this errant proposed theory, commitment to it would require rolling back commitments to earlier theory of electrons, and the burden of proof would lay on those instead. Put simply, the dark matter theory would just become an electron theory instead and would be quickly rejected. Separate from her thoughts on unobservables, she smartly points out that an observable distinction is no less helpful to the truth of theories. She refers to the demonology “science” of the 17th century in the hunting of witches, wherein it was construed that witches existed and would have certain properties and that are such and such. The ability to observe a candidate for what could be a witch gave observers the impression that a true conclusion could be reached as to whether a person was in fact a witch, never challenging the theory that witches indeed exist. So, the theory that witches exist is based on the fact that witches are observable, which they would be if there was one in front of you, but there isn’t (Contessa, 2006).
Another of my favorite arguments (because of its simplicity) against Fraassen is from Anjan Chakravarty. “[I]t is widely held, not least by realists, that even many of our best scientific theories are likely false, strictly speaking, hence the importance of the notion that theories may be “close to” the truth (that is, approximately true) even though they are false” (Chakravarty, 2011). So, if it is true that realists believe their theories are true while still acknowledging that they are likely false, it implies that the word believe means something to the realist other than blind faith. In my mind, the two ideas are not mutually exclusive, but instead that the belief is contingent on new findings, or proofs, or even an inference to a better explanation.
3.3 My View
I have very little to add to this conversation, save for the few opinions I’ve taken the liberty to pepper inline above. The main reason I have nothing more to add is that the debate does seem to be more semantics than anything else. I agree with the realists that there should be space for unobservables within theories posited, with the understanding that there must be at least some empirical data that supports the claim. Whether the theory presented is held as “mostly true” or just “empirically adequate” seems to make no difference to me, since true theories persist the tests of time and false theories do not. (This idea is inspired by Putnam’s “no miracles argument”, of which I read about in Chakravartty, 2017). There are two remaining opinions regarding this debate that I believe put the issue to rest. The first is from philosopher Helen Longino regarding the nature of the theory-observation distinction, and the second is from Bas van Fraassen himself. First Longino’s take: “An antirealist… holds… that science aims not at truth but at empirical adequacy. A realist… holds that the success of science can only be explained by the literal truth of scientific theories. These positions are… about the semantics of theories, the aims of science, and what one’s epistemological theory does and does not license.” (Longino, 2002, p. 141). This explanation seems accurate to me, and while Fraassen (1980) in his introduction timidly introduces the term “constructive empiricism” as an alternative viewpoint, calling it a “momentary name” (p. 5) he has an interesting postscript in the final chapter of The Scientific Image; “But now I must report on the reasons which have converted me to total belief in scientific realism. This change of mind was a sudden occurrence, taking me unawares when I was reading Aquinas” (p. 204). Being that this writing was over forty years ago, it is possible that Fraassen would no longer admit this, as he continues to be known as the constructive empiricist of note today. Or perhaps he was acknowledging at the time that distinctions between constructive empiricism and realism were semantic in nature, and only after the sweeping response to the idea of constructive empiricism in the philosophy community did he fully identify with the position. Despite my best efforts to find an answer to that question, I am still left without an answer. But just like in the sciences, many things are both/and, and so the answer could be “all of the above”. The answer to the initial question of whether scientists can posit the existence of unobservable entities, I would say the answer is clearly yes. For, no matter which discipline of philosophy we’re operating under, it is permissible. The distinction, as we’ve touched on several times, is the difference in epistemology only.
References
- Special thanks to Ben Horne for this helpful insight
Boyd, Nora Mills and James Bogen, “Theory and Observation in Science”, The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/archives/win2021/entries/science-theory-observation/
Chakravartty, Anjan, “Scientific Realism”, The Stanford Encyclopedia of Philosophy (Summer 2017 Edition), Edward N. Zalta (ed.), https://plato.stanford.edu/archives/sum2017/entries/scientific-realism/
Van Fraassen, Bas C, “The Scientific Image”, Oxford: New York: Clarendon Press, Oxford University Press, 1980
Gabriele Contessa, “Constructive empiricism, observability and three kinds of ontological commitment”, Studies in History and Philosophy of Science Part A, Volume 37, Issue 3, 2006, Pages 454-468, ISSN 0039-3681, https://doi.org/10.1016/j.shpsa.2006.06.007.
Kukla, Andre, “The Theory-Observation Distinction” The Philosophical Review, vol. 105, no. 2, 1996, pp. 173–230. JSTOR, https://doi.org/10.2307/2185717. Accessed 25 Oct. 2022.
Longino, Helen E, “The Fate of Knowledge”, Princeton University Press, 2001
Maxwell, Grover, “The Ontological Status of Theoretical Entities”, Scientific Explanation, Space, and Time, Vol 3, Minnesota Studies in the Philosophy of Science (Minneapolis: University of Minnesota Press, 1962), 3-15
McMullin, Ernan, “Van Fraassen’s Unappreciated Realism”, Philosophy of Science (2003), 70 (3):455-478
Collection of Found Objects 
Leave a comment