Understanding, Explanation, and Scientific Knowledge

In the past decade, both epistemologists and philosophers of science have taken up the issue of how we understand empirical phenomena. As epistemologists have expanded their purview beyond knowledge, other cognitively valuable achievements such as understanding have become a natural area of study. Understanding also fits seamlessly with the skills and know-how exhibited in scientific practices—most notably explaining, modeling, and idealization—that have captured the attention of many contemporary philosophers of science. Frequently, the foil to both projects has been a “Received View” of understanding, suggested (but not developed) by philosophers of science such as Carl Hempel, Wesley Salmon, Peter Achinstein, and James Woodward. Often the new “friends of understanding” find some fault with the Received View as the impetus for their own positions.

In this book, I refine, systematize, and defend the Received View’s guiding idea—that understanding is knowledge of an explanation. My refinements to the Received View are threefold: a pluralistic stance with respect to the kinds of explanations that provide understanding, an emphasis on scientific knowledge of an explanation as the source of understanding, and an account of how understanding admits of degrees. These refinements undercut much of the motivation for recent philosophical discussions about understanding. However, far from being an exclusively critical work, the book also clarifies and answers the most central questions in this burgeoning field of philosophical research:

• What kinds of cognitive abilities are involved in understanding?
• What is the relationship between the understanding that explanations provide and the understanding that experts have of broader subject matters?
• Can there be understanding without explanation?
• How can one understand something on the basis of falsehoods?
• Is understanding a species of knowledge?
• What is the value of understanding?

In answering these questions, I employ a philosophical methodology that integrates ideas and approaches from epistemology and philosophy of science. Case studies from physics and biology, as well as more general points about experimental design and causal inference, are used to adjudicate between different epistemological theses. I also use more conventional philosophical methods, such as linguistic analysis and argument reconstruction. The result is a model of how a philosophical account of understanding should defer to scientific practice.

OUTLINE OF CHAPTERS:

1. The Philosophy of Understanding

I begin by clarifying different kinds of understanding, and identify understanding-why as my chief target of analysis. I also restrict myself to the understanding of empirical phenomena. This kind of understanding is contrasted with linguistic understanding, understanding-that, and understanding-who, -where, -what, and –when, as well as understanding in non-empirical domains, e.g. mathematics.
I also discuss how an oft-overlooked dimension of understanding—that it admits of degrees—should alter the way that philosophers approach the concept(s) of understanding. Specifically, I derive several non-comparative concepts of understanding(ideal understanding, and intermediate or “outright” understanding) from analyses of the concepts better understanding and minimal understanding. Specifically, our understanding improves in two different but complementary ways: (a) when we grasp more correct explanations of the phenomenon of interest, and (b) when our grasp of those explanations bears closer resemblance to scientific knowledge. I discuss how this account relates to the Received View, the challenges to the latter, and briefly outline how subsequent chapters in the book respond to these challenges.

2. An Illustration: Bjorken Scaling

In this chapter, I offer a detailed case study to illustrate how my account of understanding tracks with scientific practice. Specifically, I use this account of understanding to explain the reception of James Bjorken’s model of scaling by the broader physics community in the late 1960s and early 1970s. The historical episode is interesting because Bjorken’s explanation of scaling initially did not provide understanding to other physicists, but was subsequently deemed intelligible when Richard Feynman provided a physical interpretation that led to experimental tests that vindicated Bjorken’s model. In the process, I offer a general account of what scientific knowledge of an explanation entails. Finally, I argue that other philosophical models of scientific understanding (Batterman’s and De Regt’s) are best construed as limiting cases of my more general model.

3. Understanding and Ability

Perhaps the most prominent objection to the Received View is that understanding involves “special abilities” that explanatory knowledge does not require. I argue that this objection does not bear on my view. Specifically, both the Received View and the purveyors of this “Ability Objection” have treated understanding as a categorical status. By contrast, I treat understanding as admitting of degrees, so I can split the difference. In line with the Received View, some instances of modest understanding exhibit modest abilities. In line with the Received View’s critics, understanding improves in proportion to the abilities one manifests. The crucial caveat is that these abilities must play some role in scientific practices that furnish explanatory knowledge. Specifically, scientists’ knowledge of an explanation is typically the result of: (a) considering alternative explanations of a phenomenon, (b) comparing these explanations on the basis of scientific evidence and scientific methods (paradigmatically empirical tests), and (c) forming beliefs about those explanations on the basis of these comparisons.

I then show how three prominent views on understanding and ability—proffered by Duncan Pritchard, Alison Hills, and Stephen Grimm—are defensible only insofar as they respect this caveat. Consequently, we need no special abilities beyond those found in scientific practice, nor do we need a special kind of mental state, such as “grasping,” that is characteristic of understanding. Rather, the only mental states required are those that approximate scientific knowledge closely enough for the purposes at hand.

4. Robust Objectual Understanding

Another prominent objection to the Received View is that its prizing of explanatory understanding misses out on a more fundamental type of so-called “objectual understanding.” Objectual understanding is most readily associated with an expert’s understanding of a subject matter, e.g. Einstein’s understanding of physics. However, exactly what objectual understanding is, and how it challenges those views that prioritize explanatory understanding, require clarification. The most popular view holds that objectual understanding requires more than explanatory understanding. I call this robust objectualism. Robust objectualists typically argue for their position by comparing objectual understanding of one subject matter to shallow explanatory understanding of a very specific phenomenon. I argue that a valid argument for robust objectualism would instead require a comparison with deep explanatory understanding of a very general phenomenon. Once this comparison is in place, the arguments for robust objectualism are rendered unsound.

5. Austere Objectual Understanding

However, some have also argued that objectual understanding requires less than explanatory understanding. I call this austere objectualism. Austere objectualists challenge the Received View by denying that understanding requires explanation. I show that many of the examples used to motivate this view rest on agents possessing information that is a useful stepping-stone to acquiring explanatory knowledge. A notable exception is Jonathan Kvanvig’s example of the understanding we can have of non-deterministic quantum systems. While Kvanvig claims that we can understand such systems without having explanations of how they behave, I use the literature on non-deterministic explanations to undermine those claims. Consequently, Kvanvig’s example is an instance of explanatory understanding—not austere objectual understanding.

6. Objectualism Demystified

This chapter takes stock of the previous two chapters, and shows how to derive objectual understanding from explanatory understanding. On this view, which I call objectual nominalism, objectual understanding is just having lots of explanatory understanding about important facts about a subject matter, where what counts as important is largely a non-epistemic, contextual matter. I rehearse how objectual nominalism avoids the pitfalls of the objectualist approaches from the previous two chapters, and anticipate some objections to my view.

7. Understanding Without Explanation?

Much like the austere objectualists in Chapter 5, Peter Lipton has argued that understanding can exist in the absence of explanation. I raise similar challenges to Lipton’s view. Specifically, Lipton does not discuss how understanding admits of degrees. I argue that all of Lipton's examples are consistent with the idea that understanding improves as we gain scientific knowledge of more correct explanations. I also argue that this approach motivates a central assumption in Lipton’s argument. Hence, there can be understanding without explanation, but only because such understanding plays some auxiliary role in scientific knowledge of an explanation.

8. Understanding and True Belief

Yet another prominent objection to the Received View is that one can understand why something is the case on the basis of false propositions. Since knowledge entails truth, understanding is alleged not to be a species of knowledge. This, of course, runs counter to the Received View.
This objection to the Received View comes in the form of two arguments. First, past scientific theories, such as Copernican astronomy, are false, yet provided some understanding of their target phenomena. As I show, many of these examples involve approximatetly true explanations, and where they do not, claims about the understanding achieved need not be granted.

The second argument against the Received View holds that idealizations deliberately misrepresent, yet they frequently advance our understanding. In response, I argue that scientists do not believe that idealizations are true, but still accept them for the purposes of discriminating between relevant and irrelevant explanatory factors. However, this discrimination frequently results in explanatory knowledge. I illustrate this with a detailed discussion of the derivation of the ideal gas law from statistical mechanics.

9. Understanding and Luck

In epistemology, the most distinctive argument against the Received View is that understanding can be “lucky” or tolerant of Gettier cases in a way that knowledge cannot. I first raise objections to the current arguments, offered primarily by Pritchard and Kvanvig. Specifically, I claim that they conflate knowledge-undermining luck with a more innocuous kind of luck that both knowledge and understanding tolerate. I then argue that because my view of understanding admits of degrees, I can grant that there is some lucky understanding. However, I then show that scientists go to great lengths to ensure that their explanatory commitments are not the product of luck, and that their understanding improves when they adopt these luck-reducing measures. Using general discussions about experimental design and statistical inference, as well as a case study on the discovery of the bacterial causes of ulcers, I argue that this anti-luck norm governs many scientific practices.

10. The Value of Understanding

I argue that understanding is more valuable than that which falls short of it, but only in degree. I argue against the stronger view, endorsed by Pritchard and Kvanvig, that the value of understanding and the value of that which falls short of understanding differ not only in degree, but also in kind. I then show that the very motivations that would lead one to seek this stronger value-claim are mistaken. I conclude the chapter with broader reflections about the book as a whole, arguing that a common theme has been a critical stance towards philosophical theorizing that strays too far from descriptions of scientific practice.