A Response to Emery's Case for a Circularity Problem of Humean Laws
Emery begins her paper by defending the following principle:
“The pattern-explanation principle — initial version. We ought not leave a well-established pattern unexplained, even if the only way to explain that pattern is to introduce some type of entity that is metaphysically weird or novel” (3).
She goes about defending this principle by way of three key examples in history of
scientific inquiry. The first was when scientists had discovered the phenomenon of beta decay. As Emery puts it: “The total energy and momentum of the observed particles before the decay was not equal to the total energy and momentum of observed particles after the decay” (3). As this appeared to violate the law of conservation of momentum, physicists felt compelled to try to explain the phenomena. Two theories were proposed, one of them being Wolfgang Pauli’s neutrino, which, as Emery notes, was certainly a metaphysically novel entity at the time. For various reasons, historian of physics Laurie M. Brown even believes it would have even been considered “abhorrent” at the time. Despite all of this, the neutrino came to be widely accepted as an explanation when the alternative explanation was disproven (4). Presumably, this supports the principle above by showing that we ought to be willing to introduce new entities to explain when necessary.
The second example was when scientists were attempting to explain electromagnetic phenomena. A theory that they “had to be explained by some kind of action at a distance, similar to the way that, at the time, gravitational phenomena was thought to be explained” was put forward (5). However, a scientist named Faraday provided evidence against this argument and, in the lack of additional competing explanations, introduced a novel metaphysical entity: electromagnetic fields (5). Unlike the neutrino, he offered two potential explanations of what these might entail, and another scientist proposed another one. Crucially, both believed their explanations to be “tentative,” and they and their followers were willing to introduce an additional entity “given that [the electromagnetic phenomena] could not be accounted for in terms of any other, previously accepted entities” (6). This last part is crucial: since current entities did not sufficiently explain, the scientists felt it was required of them to introduce a new one, however unsure they were about its nature.
The third example was dark energy. This one is rather straightforward: Scientists learned that “the universe is expanding and accelerating,” and have “given the solution a name,” namely “dark energy” (6). As Emery notes, here we introduce an entity “about which scientists know absolutely nothing” because “without that entity there would be no explanation available for a well-established pattern in the data” (6). The introduction of this entity is meant to show a final way in which the pattern-explanation principle seems abundantly clear to be the case: even in spite of extremely limited knowledge, we feel we must posit something to explain a given well-established pattern. As Emery puts it, a plausible account of why scientists are so comfortable introducing novel entities at high metaphysical cost is that “because without such entities, well-established patterns in the data would go unexplained” (7). This is the claim on which the initial pattern-explanation principle rests.
When determining what the metaphysical equivalent of this principle might be, Emery notes that we have to identify what type of explanatory burden the principle necessitates. She argues that the examples show that we are required to give a “metaphysically robust explanation—an explanation in which the explanans identifies the reason why the explanandum occurred” (8). One might wonder what the nature of this reason might look like. Emery claims that “paradigm cases” of these explanations “identify the cause or ground of the explanans” (8). Thus, the final version of the pattern-explanation principle, she claims, should be as follows:
“The pattern-explanation principle — final version. We ought not leave a well-established pattern without a metaphysically robust explanation, even if the only way to explain that pattern is to introduce some type of entity that is metaphysically weird or novel” (3).
However, it is crucial to her argument that a metaphysically robust explanation need not be one we understand. She argues that as these entities can me metaphysically ‘weird’ or ‘novel’, “Something other than mere understanding must be required in order to discharge the explanatory burden these patterns create” (8). She defends this claim by noting that “sometimes the reason why something occurs is beyond our understanding and sometimes our understanding can lead us astray with respect to the reason why something occurs” (8). Essentially, the argument is that the responsibility to posit metaphysically weird entities to explain phenomena where other explanations are not epistemically viable is in part due to the fact that this means there is a strong likelihood our current list of metaphysical entities cannot sufficiently explain a phenomena. She argues that this “explanatory burden generated by such patterns is so substantial that it warrants the introduction of metaphysically weird or novel entities in order to establish that there is some reason why that pattern occurs, even if, all else being equal, we would prefer to keep such entities out of our metaphysics” (8).
The argument about why the principle that constrains theory choice in physics should move us towards accepting a similar principle that constrains theory choice in metaphysics is straightforward. Emery notes that is widely accepted that our best theory choice in metaphysics should be informed by the content of our best scientific theories, and argues that “one can have no good reason for thinking that the content of our best scientific theories should constrain theory choice in metaphysics if one does not also think that standard scientific practice constrains theory choice in metaphysics as well” (10).
Emery argues that while questions about the nature of governing laws abound, the pattern-explanation principle tells us that regardless of the answers, we should not resist the governing conception of laws unless “alternative explanations of the patterns in the data” (e.g. competing non-Humean accounts of laws) are developed (12). In other words, we need an explanation of well-established patterns that is metaphysically robust: one that establishes the cause or ground of the explanans explaining the explandum. And right now, the best conception we have of this is a governing account of laws.
From this, Emery argues that two key strategies characteristic of Humean accounts of laws will not work. The no-explanation strategy which argues that “There is no explanation for the fact that f = ma in all observed cases” is insufficient because it violates the final version of the pattern-explanation principle (12). The pattern subsumption strategy which argues that “The explanation of the fact that f = ma in all observed cases is that f = ma in all cases” does not work because it fails to explain what grounds the fact that f = ma is true in all cases (12). Since her argument requires that we take up a governing conception of laws, it requires that we take up a conception of laws that is anti-Humean.
My objection stems from her claim that “Understanding, however, is neither necessary nor sufficient for a metaphysically robust explanation” (8). To claim that whenever it is the case that the explanation of a phenomena is beyond our understanding, we are obligated to posit a novel entity if necessary doesn’t seem to follow from the examples. Just because there *could* be an explanation beyond our understanding does not mean we are obligated to posit one. Plausibly, we can just assume a position of ignorance. How are we supposed to go about determining when it is required of us to posit explanations beyond our understanding, rather than simply claiming ignorance?
From the examples, one might say that it is required of us to posit such explanations when all alternative explanations within our epistemic sphere have been shown to fail. This was the case for widespread acceptance of the neutrino and the electromagnetic field. In terms of ‘dark energy’, we have little to no idea of what the nature of this entity would even be like. It is not ‘among explanations’, it is simply a posited entity that is meant to fill an explanatory gap. What seems strange to me is that we have minimal understanding of how, exactly, it would fill this gap or what this gap even looks like. There thus seems minimal reason to be obligated to posit an entity to fill the gap.
So, working off the examples given, there are two potential reasons that we ought to fill in an explanatory gap with an entity, even if metaphysically weird or novel: 1) We have strong reason to believe alternative explanations of a well-established pattern within our epistemic sphere have failed and 2) We have no explanations of a well-established pattern within our epistemic sphere, largely due to a blindness of the nature of the problem at hand.
I have already provided my argument as to why the second option fails to be a situation in which we ought to posit metaphysically weird or novel entities even at the scientific level: it is a situation we simply know very little about, and thus a claim about how we should attempt to explain the situation is largely unwarranted. Although we have a well-established pattern, we know very little about the *nature* of the pattern. The reason we have no alternative theories is that we don’t have a precise idea of what we are trying to explain. In the case of the neutrino, we know that momentum and energy are being lost, concepts we are familiar with. In the case of the electromagnetic field, we had some awareness of the *nature* of the phenomena we were trying to explain that allowed us to effectively test how they operate. If we know next to nothing about what the forces causing expansion look like, it seems unclear that we are justified in arguing we are obligated to posit a metaphysically weird entity to explain them. We can be no more confident in a metaphysically novel entity with little descriptions than in a brute fact.
If we were to reject a Humean account of laws through reason #1, we would have to reject it on the grounds that we have strong reason to believe such an account has failed. From the examples given, we can only extrapolate that these reasons should be given in the form of empirical evidence. It is unclear how one would even go about empirically debunking a Humean conception of laws, so the attempted rejection fails here as well.
In summary, it appears to me that we should revise the initial pattern-explanation principle as such:
The initial pattern explanation principle (that I think the examples actually defend): We ought to not leave a well-established pattern—that we understand enough about to empirically test epistemically reasonable and accessible alternative explanations of—unexplained, even if the only way in which one can explain some well-established pattern that we understand enough about to empirically test epistemically reasonable and accessible alternative explanations of is to introduce a type of entity that is metaphysically weird or novel.
This emphasizes the point that the examples seem to only recommend pushing forward a novel metaphysical entity once all alternative empirically testable explanations are rejected. It also rejects the notion that one is obligated in scientific theory choice to introduce novel metaphysical entities for patterns whose nature we have extremely limited understanding of.
If we were to “finalize” this principle using Emery’s methods, it might look like something close to the following:
We ought not leave a well-established pattern—that we understand enough about to empirically test epistemically reasonable and accessible alternative explanations of—without a metaphysically robust explanation, even if the only way in which one can explain some well-established pattern—that we understand enough about to empirically test epistemically reasonable and accessible alternative explanations of—is to introduce a type of entity that is metaphysically weird or novel.
The question of whether laws govern is not one we understand enough about to empirically test epistemically reasonable and accessible alternative explanations of, and we have limited access to what the nature of these governing laws would *be* exactly, so the argument does not go through.
Perhaps Emery would respond that in the cases of neutrinos and electromagnetic fields, these entities were justifiably posited even prior to there being alternative explanations. The fact that competing alternatives did exist is irrelevant to whether the examples defend the initial pattern-explanation principle. The fact is that these explanations with novel metaphysical entities were produced and justifiably so: they were legitimate explanations of a well-established pattern. Additionally, regardless of the nature of competing theories, the project of positing novel metaphysical entities, even prior to rejections of alternative theories, has a strong track record of being a successful way of predicting scientific explanation well, as the examples show. This is relevant because it provides evidence for the claim that, when doing physics, we ought to posit entities when no other explanations are available.
Where this response seems to fall short is that it proves too little: it merely shows that when faced with a well-established pattern, we are justified in positing a metaphysically novel entity. The claim we ought to based of historical precedent, which is the claim Emery makes, still seems to rely on this track record of empirical success. We have no way of empirically testing the nature of laws yet, so the shift of the ought claim from physics to metaphysics is unjustified.
Comments
Post a Comment