Science and Analogy – A Reply to Glenn Statile’s Analogy and the Integrity of Science
By
Lowell Kleiman
One of the more perplexing aspects of modern science is how scientists know what they are talking about. How do they know that there are tiny particles that compose all the things we commonly see and feel especially since the bits of which they speak are imperceptible and most likely to remain so? How can they be sure of electrons, protons and neutrons, let alone bosons, muons and gravitons? What about up, down and charmed quarks, or strings, the ultimate denizens of the physical world? What sense can such theoretical talk make?
The explanation favored by Glenn Statile, which he shares with others in a tradition dating back to Descartes, is analogy. We learn about the objects of science through comparisons with the familiar, planetary motion like falling apples, electromagnetic conduction like water through a pipe, gases like collections of tiny billiard balls. Of course, the analogy must be honed and sharpened – molecules of water lack the colors and printed numbers we observe on a pool table, but short of that and other irrelevancies, the analogies are apt and illuminating.
Few in science would disagree. What then is the issue? Statile believes that analogies are more than just useful devices for understanding science; they are “necessary conditions for the justification” of scientific theories in that hypotheses are wrong “if they cannot in some way be likened to, or justified by, a scientific model or analogy.”[1] Statile calls this thesis the “strong” view and I agree. Unless gases are thought of on the analogy of collections of tiny elastic billiard ball-like objects, the laws espoused by Boyle and later theorists would not work. Soft, little gummy bears would behave erratically under conditions of temperature and pressure, and would fail to support the predictive power of the kinetic theory of gases.
Why is the ‘strong’ view controversial? Those of us who believe in the necessity of “semantic augmentation,” the need for an analogical interpretation of an otherwise abstract scientific formalism, face the following dilemma: for any explanatory analogy in science, either the analogy is independently confirmable or not. If it is independently confirmable, then the analogy is eliminable, at least in principle, and fails to be necessary as the strong view claims; if it fails to be independently confirmable then no theory is justified. So either the strong view is false, or science is unknowable. Not a happy choice for Statile or any of us who support his view.
What answers have been given? Descartes accepted the strong view proposing that both sides of the analogy be symmetrical – that the imperceptibles be thought of as familiar objects that are very far away or just too small to see with the naked eye. But modern physical theory includes objects like strings that are in principle unobservable. Descartes’ solution is insufficient.
The positivists rejected the strong view arguing for an empirical method of justification based on observed relations – in the study of gases, that pressure and volume vary indirectly – along with a formal method of reducing the theoretical to the observable. The program depended on a verificationist theory of meaning among other considerations which have yet to be established.
By contrast, modern pragmatists accept analogy but reject independent confirmation as too strong. They opt for inference to the best explanation as a way of avoiding the troubles. If the gas laws work under the interpretation of tiny colorless billiard balls, then so be it. Problem is even gummy bears are elastic in the physicists sense – up to a point. Unless we can distinguish in advance one analogy from another, we can’t be certain that enquiry, even for practical purposes, is at an end. Short of that we can’t be sure that the gas laws, or any laws, hold.
What about the realist’s idea that the entities on the theoretical side of the analogy are part of the cause of our believing in them? That’s okay provided we are sure that there are some such entities. But that’s the problem – without independent confirmation how do we know what’s there?
Where does this leave us? Must we turn to skepticism, abandon the strong view, or just lower our standards? As Statile notes, “After all, as we all know, science itself, correctly understood, is not engaged in the business of proof.”[2]
The issue, however, is not whether an hypothesis is proven but whether analogy is necessary in the justification of theory. If the integrity of science is to be upheld, more work is needed to support this view.
Lowell Kleiman
Suffolk County Community College -- SUNY
Selden, NY 11784
2
[1] Glenn Statile, Analogy and the Integrity of Science, delivered at St. John’s University, Oct. 18, 2008.
[2] Statile, page 9.