I get questions regularly about the bizarre nature of contemporary physics. I am sure practicing physicists with PhDs get these more regularly than I, yet I occupy an interesting and rare position in the academic disciplinary landscape: I’ve studied science, particularly physics, into the graduate level, and I am actively developing my expertise in the history and philosophy of science, particularly physics, as well as being a lifelong student of more traditional philosophy (e.g., analytic, contemporary, and Eastern). The question most regularly asked of late has been: What are physicists talking about with all of this “non-verifiable” theory; it sounds like philosophy? By this, they mean the fact that there is this apparent post-empirical turn, and the lack of requirement of empirical data to substantiate proposed theory. I’d like to spend some length explaining my thoughts on this, including a suggestion to all practicing scientists, regardless of discipline.
What’s prompting these questions are articles like those by NPR (click here) and physicists George Ellis and Joe Silk (click here). The sum of the concern and conflict is stated nicely by those two physicists, who say:
This year, debates in physics circles took a worrying turn. Faced with difficulties in applying fundamental theories to the observed Universe, some researchers called for a change in how theoretical physics is done. They began to argue — explicitly — that if a theory is sufficiently elegant and explanatory, it need not be tested experimentally, breaking with centuries of philosophical tradition of defining scientific knowledge as empirical. We disagree.
Aside from that bit about the “centuries of philosophical tradition” —an erroneous phrase—, this accurately depicts the clash. Particularly, in their article, they comment on the fact that so many physicists wish to lower the standard of what passes for accepted theory, because the options for understanding the most fundamental level of the physical world, the quantum world, admit no mode of direct or indirect observation for confirmation, i.e., they admit no empirical grounds for determining the truth of a statement, such as those asserted by a theory. These theories might be referred to as the “many worlds” theory or interpretation and “string theory.” These have been at loggerheads for a little while now, and tension is mounting. Physicists, who want to be viewed as those in the discipline which can truly know that nature of the physical universe, and limitations within the discipline to do so are admissions of impotence. Hence, we have a political and economic situation where, even though neither of these interpretations might turn out to reflect reality in anyway, professionals feel compelled to have answers.
If physicists and scientists, generally, were a bit more honest, a bit more interested in truth, a bit less interested in cashing in on that next grant, a little less interested in that next, space-consuming, time-wasting, hardly-worth-writing article, and a bit less interested in the social and political ramifications of being viewed as have access to “secret knowledge” of the universe, they’d come to a few conclusions. One might be that they’d say, “Well! We appear to have come to a situation in which it seems the content of the answers to are questions transcend our methods’ access, and so maybe there is no way for us to determine whether many worlds or string theory, or what have you, are true.” One conclusion they would come to is that there are two kinds of science, which might be broken down in to further subcategories, but that perform two separate functions: one is to explain the natural world for the sake of understanding, the other is to increase the human capacity to manipulate the world around us. Many scientists, particularly physicists, will not like that distinction, claiming that sciences can do both. However, there is a very simple fact of the matter, which is impossible to refute, otherwise, something very bizarre would happen, and that fact is that data and theory are never married. Pierre Duhem, the great physicist, mathematician, historian, medical enthusiast, and natural philosopher, pointed this out in The Aim and Structure of Physical Theory and elsewhere. The bizarre thing that would happen if the contrary were true, if data and theory could be irrefutably conjoined, they would not need to be conjoined, but that the data would, itself, present the theory to the individual reviewing the data set; the data set would spring for a fully formed theory, as Athena from the head of Zeus. There is a simple takeaway from this, and it comprises the suggestion I have for scientists: the separation of theory and data makes a primary distinction between two forms of science, a science with a pragmatic philosophical outlook and a second that focuses on understanding. That is, the separation between data and theory, ever being asunder, means that a pragmatic science —for our purposes, a science oriented in the interest of manipulating the physical world— acknowledges its impotence in capacity to explain, and only focuses on the truth that happens. In other words, a pragmatic approach to science means accepting that the current theory or theories in play may or may not be correct, but produce approximately the desired outcome in manipulation. For instance, the whole the foundational theory of chemistry might be wrong, but still capable of providing some ability to successfully manipulate the physical world. The ahistorical minded individual might not be in any way aware of the fact that phlogiston theory had quite bit of success in explaining how to turn ores into usable metals, or that Romans built their great aqueducts with Aristotle’s physics and thinking that “things of a like tend toward their place,” or that medical theory employing the four humours had success in treating various conditions. With the same hubris that scientists have today, and which seems to be a natural foible in the human condition across all times, those practitioners and researchers thought they had the Truth. Instead, I think it is time science developed a little wisdom and took a new tact.
The important part of this new wisdom means acknowledging that the “understanding” portion of science, where narratives are supplied to explain what it is that scientists qua humans think about what it is that data might be saying and the structure of Nature. In the minds of practicing scientists, it means an immense downgrade in status of what the sciences do. In reality, as reflect by my opinion, I think it is a massive upgrade for science, being that the truth supplied by Duhem indicates that manipulative knowledge of the world tells us nothing, as it understands nothing. Human judgment and story-telling are required to attempt making sense of data. The fact that scientific theory entails positing semi-ontological entities means that we’ve moved away from the world of sensuous experience, and are grasping at making of sense of the world through things that may or may not be there. Much of science today should be viewed in the way it has been in those “centuries” prior that George Ellis and Joe Silk referred to, i.e., as a creative story-telling endeavor, where we employ empirically inspired thought, speculative reason, deductive arguments, inductive arguments, and, other creative interpretations —though I view all interpretations as created “creatively” ex nihilo, and so the redundancy is for emphasis only.
What this means for the scientist interested in furnishing dogmatic disciplines (e.g., pharmacology, medicine, engineering, etc.) with further capacities (technological prowess, we’ll say), there is no stock placed in whether an understanding can be constructed on the basis of what pragmatic truths they are able to conjure. No such investment is needed. Regardless of the story behind the substance that explains what insulin does, it generally does something —it makes a truth happen, and that is its cash-value. Going beyond that, going into story-telling about the world, and creating a narrative about how the world works —that’s philosophy. The further one enters into such narrative construction, the further one moves into the realm of philosophy. View science in these terms is what needs to happen, in order for science to regain its healthy psychological complexion, and, until it does, Ellis and others are going to have to continually deal with scientists who are trying to do metaphysics within the domain of science. And that’s precisely why I get the questions that I do. It is abundantly clear to the reader of popular-culture physics books that what scientists are doing is more like mythology, in some cases, philosophy in others, sometimes both.
When it comes to metaphysics, we are talking about a great many things, but we certainly aren’t talking about the practice of science. Metaphysics is a multifaceted word that etymologically means after physics (physis in Greek means Nature), or beyond physics, or, stretched a bit, undergirding physics. When I teach my undergraduate courses, I give a collage of depictions, not a strict definition, and these depiction include a contrast with ontology (particular entities of the immediate world, such as gravity qua law, whereas metaphysics is general and more abstract, as in the proposition that the world is governed by laws); it includes the remark that metaphysics is the speculative realm of the study of existence; it includes the consideration that metaphysics deals with establishing an understanding of the conditions that would make existence on the ontological level possible; and it is the regime in which limited circumstantial positive knowledge is possible, but nothing more. These are reflected in thoughts by St. Thomas Aquinas who, in his Summa Theologica, said that the metaphysical realm is the realm of religion. When it comes to complete underdetermination of claims, as admitted by the physicists arguing about the many worlds interpretation and string theory, it must be acknowledge that the spirit of pragmatic science has left the room. None of this is to say that physicists should not do this, but it is to say that they should admit what they are doing: philosophy of physics. Moreover, they need to also admit the kinds of truth criteria differ from that of pragmatic science. In particular, within pragmatic science, truth happens, whereas it certainly does not in the non-pragmatic, philosophical variety of science. To be clear, the value in this latter variety of science is that it can help induce new truths to happen within the pragmatic domain of science. As any historian of physics knows, the sudden interest in biology taken by physicists in the question of life and material transmission of heritable information lead to a period of speculative philosophy of science (see Max Delbrück’s involvement in genetics and Schrӧdinger’s inquiry into the question of what biological life is) illustrates this nicely.
To conclude, I think scientists need to be clear on what the nature of their scientific activity is, and the objective of interest-dependence, as Bas van Fraasen would put it, determines the nature of that scientific activity. Dragging metaphysics into science has only every result in new scientific methodologies in cases where the questions became precise enough; it has not, of its own ability, ever made truth happen —and that’s the only definition of truth that has ever really worked in producing technologies (i.e., the substance of scientific progress), that truth is what happens.
 Erroneous because it would take a hell of a lot of historical analysis and philosophical arguing to claim that: 1) science, as we understand it today, existed prior to the 1850’s, and 2) that the standard of physics’ theory production has ever relied only upon empirical information, as we know that it absolute played no role in some theory production (see Descartes, Leibniz, and similar early natural philosophers). Perhaps the most erroneous aspect of the phrase is the fact that physicists accidentally air their obsession with their disciplines hierarchical disciplinary placement among other disciplines and their physics chauvinism, in that they maintain that physics provides the standard for what a science is, meanwhile, most sciences (cosmology, astronomy, evolutionary biology, geology, etc.) have few or no experimental methodologies or invasive procedures to make determinations about the natural world. I therefore view this mark as rhetoric for gusto.
 Please refer to the observable/non-observable debate in the philosophy of science. It centers in the fact that some entities within ontology seem to be strictly observable through the sense, while others are indirectly observable, and others not observable at all. Making matters worse, it is not entirely clear where to draw the line in simple situations, such as looking through a telescope at an object that is also visible to the naked eye.
 The chasm between ontology and epistemology leaves one with the question of the mind could ever hope to guess the nature of ontology. Epistemology, in my view, lacking a universal and absolute standard, entails making ideas and trying to fit the framework of ideas to the world through various methods. There simply is no certification process known to man that would verify any linguistic statement, and so it seems impossible to know whether any idea or framework ever (!!!) has accurately and totally reflected ontology. For pragmatic science, all one can say is that, whether a theory is right or wrong, truth happens, in the sense that we got approximately result X from thinking approximately Y, and performing approximately Z; the genuine reason as to why the truth happened remains elusive.
 I have to cite Steven L. Goldman’s brilliantly constructed lecture series published by the Teaching Company, called “What Scientists Know and How They Know It,” wherein he remarks that, one fine day, the scientists broke away from the philosophers, only to wake the next day and begin doing their own philosophy —and poorly, at that!
3 responses to “Distinguishing between Types of Science: Unmixing Metaphysics and Pragmatic Science”
Your post reminded me of Paul Feyerabend’s response to the famous letter signed by 180+ scientists back in the 70s putting astrology on the Scientific Index of heretical studies. http://digilander.libero.it/astroitalia/cialtrones.pdf
And the bit about theory and data not lining up reminds me of this recent piece about Big Bang Inflationary models: https://www.quantamagazine.org/20150130-joint-dust-analysis-deflates-big-bang-signal/
That’s kind of funny you bring up Feyerabend, because I was thinking about Kuhn while writing.
The BICEP study is interesting, because it illustrates the sociological phenomenon of groups of scientists attempting persuade the field, through statistical manipulation, rhetoric, and who knows what behind the scenes, when it simply isn’t cost effective or economically possible to have another set of scientists produce a second set of data. That’s one of the problems with big science: good empirical science means separate scientists producing, empirically, separate data sets (to eliminate personal bias and systematic methodological error), doing so with a second set of instruments (to eliminate systematic instrumental error) and working out the correlations (to eliminate mathematical and interpretive error), etc., but you can’t do that with something like the LHC, and it’s exceedingly difficult with projects like BICEP. To an extent, I am of a mind with Fred Hoyle, when it comes to big science.
Thanks for the comment.
First, I enjoy your posts. Yes, Dr. Goldman’s class on “What Scientists Know and How They Know It” is fun and informative. Finally, after being pummeled as a denier and seeing science used as an rhetorical tool, I asked what I thought was a simple question. What is science? So far, I’ve not been satisfied with an answer.