Explanations, Narrative, and Understanding: Intersections Between Science and History

Some months ago I wrote about “The Logic of Narratives.” At the end of the post I made a big observation and asked some big questions:

Plausibly ordered evidence, presented coherently and with some through-line, and with some style (simple to complex), seems to trump all forms of argument presented, as by philosophers, in logical form. How is this so? What does it say about our minds? About logic as taught in philosophy? And finally about the power of storytelling—of history?

After reading the post, an old Loyola friend, Abe Schwab, formerly a graduate student in philosophy and now a philosophy professor at IUPU-Fort Wayne, suggested I read this essay: Trout, J.D. “Scientific Explanation and the Sense of Understanding.” Philosophy of Science 69 (June 2002): 212–233. I have now read the piece.

The article’s abstract relays that Trout will argue two points, the first being subservient to the second:

(1) “Sense of understanding is in part the routine consequence of two well-documented biases in cognitive psychology: overconfidence and hindsight.”

(2) “In light of the prevalence of [these two forms] of counterfeit understanding in the history of science, I argue that many forms of cognitive achievement do not involve a sense of understanding, and that only the truth or accuracy of an explanation make the sense of understanding a valid cue to genuine understanding.”

So we have several terms, phrases, and subjects to understand—as presented by Trout—in order to fully absorb his argument: sense of understanding, cognitive psychology, overconfidence, hindsight, history of science, science (today/present), truth, accuracy, explanation, scientific explanation, and genuine understanding. And we need to understand condition (1) above before we can understand the full argument (2).

Given that the article is only twenty pages long, it would be a tall order to both define all of these terms and also build his own argument. So Trout assumes we understand some of the above: truth, accuracy, genuine understanding, cognitive psychology, history of science, and science today (empiricism). As such, Trout focuses on explaining sense of understanding, overconfidence, hindsight, explanation, and scientific explanation.

As one reads the article, it becomes clear that Trout offers no clear line between explanation and scientific explanation. He ranges far beyond just science when he discusses sense of understanding, overconfidence, hindsight, and the psychology of understanding. And his line between examples from the history of scientific explanation and scientific explanation today is also not bright. The blurring of these two lines creates problems in Trout’s essay, but I’ll save that criticism for later.

In the body of the article Trout relays his argument across pages 213-215. He argues that “the psychological sense of understanding is just a kind of confidence, abetted by hindsight, of intellectual satisfaction that a question has been adequately answered. This sense of satisfaction is confidence that one enjoys an accurate description of the underlying causal factors sufficient…to bring about the phenomenon we are examining. but confidence is, notoriously, not an indicator of truth” (p. 213-214).

I get the point about the mere feeling of confidence and truth (or genuine understanding). But do scientists (social and/or hard) routinely relay “confidence” based on mere hindsight? I think not. Is it not true that the confidence in results obtained by scientists rests on repetition of an experiment, as well as the isolation of variable?

In the hard sciences, as I understand them, confidence in an explanation (or a narrative or a sense of understanding—i.e. results) comes from the repetition of an experiment, often with minor changes in variables such that the scientist can eliminate false explanations. Control groups are often used. The explanation aspires to such a point that if ten scientists (say physicists) ran the same data according to the same methodology, 9.5-10 of them should get the same results. This has nothing to do with the feelings or state of digestion of the physicist have nothing to do with the core of the explanation offered. The essence of the narrative should be the same.

In the social sciences, confidence derives from either experimentation or the reliance on hard data. Results and data can be both quantitative and qualitative. In history, for instance, confidence in an explanation comes from using multiple points of evidence, over time, to construct an qualitative explanation that connects as many data points as possible and is plausible (if not perfectly reproduceable by others). If ten historians looked at exactly the same data, we might see a 60-70 percent correlation in explanations. The feelings and state of digestion of each historian may, in fact, account for some core differences in explanation—in the narrative of each historian.

So I can see Trout’s point if he’s talking about historians of science and how they relay scientific explanations. But I reject Trout’s assertion if he’s talking about scientific explanation as offered by scientists over the last 100 years or so, and especially last 50 years.

Trout offers more on his thesis in the body of the article—and adds to my point about his fuzzy lines between explanation/scientific explanation and historical scientific explanation/today’s scientific explanations. After discussing examples from the history of science (Darwin, Avogadro via Perrin), Trout writes (bolds mine):

“At the moment, there is neither a satisfying formal account of explanation, nor agreement about the important informal criteria for good explanation, producing what one review casts as ‘an embarrassment for the philosophy of science‘ (Newton-Smith, 2000, 134). Current theories of explanation may leave us embarrassed, but we should be at least as embarrassed by our ‘feels-right’ diagnostic standard for the acceptance of an explanation” (p. 214).

So is Trout upset with explanation today, or in the past? Is his sense of past explanations colored by the narratives of historians of science? What present-day scientist use a “‘feels-right’ diagnostic” for their explanations? Trout does not name any, either in the body of the article or in his footnotes. Only historical figures are named. Or is Trout primarily concerned with this problem (i.e. feels-right) in the past?

What do historians of science have to say about this “embarrassment”? No historians of science are mentioned in the text or the footnotes, so I have to assume Trout has no problem with them. Trout also does not discuss those who theorize or philosophize about historical narrative (e.g. Gaddis, Fischer, Foucault, Lovejoy, Bourdieu, Jameson). Even the physicist, historian, and philosopher of science Thomas Kuhn is not mentioned (i.e. not discussions of paradigmic explanations or paradigm shifts). Kuhn’s rival Michael Polanyi is also not mentioned. Given these absences, one must assume that Trout is concerned with present-day explanations primarily—meaning that historical examples are meant to be indicative of present-day problems. Given his thesis, Trout will only be working in an area that constituting overlap between cognitive psychology and philosophy as it deals with explanation.

This overlap brings up another area I found confusing in the article. In the opening pages, even before Trout reaches his thesis, he seems to conflate ‘feeling’ [i.e. feels-right, the “affective component” (p. 213)] and ‘sensing’ (i.e. sense of understanding). The former obviously involves psychology (cognitive and otherwise), but the latter involves cognitive psychology [“psychological sense of understanding” as “confidence” (p. 213), or “subjective sense of understanding…conveyed by a psychological impression”] and traditional philosophy.

Trout hints at a traditional philosophy of the sensing-feeling distinction when he gets to “the traditional account” of understanding, as related to explanation, as based in reconstruction: “understanding requires that the individual be able to piece together bits of information in their cognitive possession.” He continues: “Reconstructive accounts of explanation agree on one point: Understanding is centrally involved in explanation, whether as an intellectual goal or as a means of unifying practice” (p. 215). In support of his traditional account Trout cites Peter Achinstein, David Lewis, Wesley Salmon, Michael Friedman, Peter Railton, Philip Kitcher, and James Woodward—all publishing papers in the 1980s and 1990s. Each of this figures “associate[s] explanation with understanding.” Layering Ronald Giere into his account of reconstruction, Trout incorporates terms and phrases like detailing, fitting (establishing fit), identifying, and detective work.

I was struck by how Trout’s accounting of traditional scientific explanation correlates with good historical thinking (e.g. pondering causation, change over time, complexity, context, and contingency) that lays the groundwork for narrative construction. Put another way, historians identify sources to (re)construct a narrative that involves fitting evidence (that could be used in contingent fashions), narrowing sources of causation from a broad context, explaining change over time, and accounting for complexity. I guess this shows how historians are social scientists as much as they are humanists. Good critical thinking translates across fields.

But two problems arose for me in these passages.

First, what of other traditional sources used to explain how one arrives at a genuine sense of understanding—how one builds a narrative of understanding without equivocating between sensing and feeling. Take for instance Aristotle’s (and Aquinas’) division of understanding into sensing, perceiving, apprehending, and judging. This way of breaking down understanding is not addressed in the text.

Although Trout uses other, non-Aristotelian examples from the history of science and philosophy, he defaults toward recent cognitive psychology to work through, or up to, a meanings of explanation and senses of understanding (pp. 216-222). He uses William Brewer (et al), William James, and Carl Hempel, as well as the aforementioned Salmon, Friedman, and Kitcher. In talking through these figures he focuses on issues like coherence (including completeness, plausibility, and consistency), objectivism in explanation (features of external objects), causal mechanisms, the “scientific world picture,” cognitive efficiency, tractability, global v. local knowledge, and the “internalization of argument.” Again, these issues—particularly the notion of a “scientific world picture”—beg me to ask: what of Khun and Polanyi? The line between cognitive psychology and epistemology gets really thin in these conversations. It would have helped if Trout had supplied some sense of his difference between the two and why that difference matters in his article.

Second, and this a major complaint that spans the entire article, but was reinforced after reading the cognitive psychology section: what of the difference between delivering or composing an explanation (scientific or otherwise) and receiving the same? From who to whom? Neither Trout nor the thinkers he cited—at least in the material provided—work hard at this distinction. It is explicitly mentioned in two spots: Brewer notes the “reader-hearer” (p. 216), and Trout addresses the “the ‘transfer of structure’…from one individual to others” (p. 222). But both discussions are very brief, done in passing.

This a major flaw in the article, haunting every discussion of the idea of explanation. Explanations are always constructed with an audience in mind, whether it is one’s self, one’s colleagues, or the general public. The author-receiver distinction should have been made early, and thoroughly explained. Aren’t sensing and feeling entirely different phenomena in relation to the deliverer and the receiver?

As for the rest of the article, particularly the discussion of hindsight and overconfidence biases (pp. 223-229), I observed no major problems and had no complaints. In the historical profession we use the term “presentism” to describe what Trout calls hindsight bias in scientific explanations. It is a problem, a fallacy, in historical explanations, particularly in relation to recent history (post-World War II). But this problem is well-known—taught to first-year graduate students and sniffed out quickly in professional settings (whether peer-reviewed articles or conference papers). As a historical fallacy, presentism is solved by instilling requisite humility in the author and the reader. That humility comes through in historical thinking when the author and reader properly acknowledge complexity, a wider scope of context, and subjectivity (i.e. perspective) in explanations.

I want to thank Abe Schwab for bringing this piece to my attention. Even if I’m dissatisfied with major parts of J.D. Trout’s article, it has caused me to think in great detail about the intersections between history and the sciences. Before reading this I had mostly only thought about overlaps in terms of evidence and a posteriori/inductive thinking. I had never given thought to intersecting issues of explanation and narration. I had always thought of scientific explanation/narration as a bare bones, stripped down affair—not subject to the fallacies that plague flowery historical exposition. Now I know differently. – TL