« No automatic DNA collection before conviction | Main | Science and/or Religion »
November 21, 2006
Scientific Methods
It has been a recurring debate among the cognescenti who study science and law as to whether there is any such thing as "the" scientific method. Among working scientists, of course, there is really no debate, since that is what they do for a living. It may very well be that there is no one METHOD associated with SCIENCE. But there are certainly METHODS that are, and that is what scientists do when they test hypotheses. Hypothesis testing, albeit using a very wide variety of paradigms, is what constitutes the scientific method (or scientific methods, if you prefer).
Well, today's blog is not the place for me to engage this debate fully, if a blog is ever quite appropriate for such matters. But a story in today's New York Times made me think of this debate, since it chronicles what it calls "the best science show on television." The show, on the Discovery Channel, is called "Mythbusters." The hosts like to "blow things up," and in so doing teach us all a lot about science. The show, as the story tells it, is not really a science show at all, since they don't teach "science." They do, however, teach the core lesson of science, which is to test hypotheses. It's an important lesson, and one that all lawyers, judges, and policymakers could benefit from learning. Now if we could just get the show to devote a few episodes to some of the forensic identification sciences, or clinical medicine, or clinical psychology/psychiatry. Unfortunately for all of us, these subjects, possibly with the exception of arson investigation, do not lend themselves to fires and explosions, and thus do not make good television. See Story Here.
November 21, 2006 | Permalink
TrackBack
TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341bfae553ef00d834fe406269e2
Listed below are links to weblogs that reference Scientific Methods:
Comments
I'm not a scientist by any stretch. Rather, I'm a lawyer, which means that I am full of opinions that no amount of empirical evidence can touch. Here's one of them. The price of generalizing that "hypothesis testing" is the common thread in scientific methodology may be accepting a very broad notion of what a "hypothesis" is and what it means to "test" one -- so broad a notion, indeed, that the generalization can become something of a bromide. That said: Yes, it is better to measure truth against empirical reality, which is refractory to the touch, than just to make up some comforting system of untestable truths (an approach that nevertheless enjoys an astonishing following these days, even among nonlawyers).
A less glamorizing generalization, often overlooked in these debates, might involve the central importance of characterization and classification in science. There will not be television shows about those scientific labors either, but they are important. Before objects or events can be measured or counted, or lawlike hypotheses about them even formulated, let alone tested, there must be at least some tacit, working definition of the class of objects or events at issue. These definitional or classificatory regimes are not always constrained to any unique form by logic or observational fact, and choosing between them can be an exercise in pragmatics, depending partly on whether one's primary interest is in articulating questions that can be answered in vaguely nomological ways by experimental or statistical inquiry, or rather in some other goal (e.g., legible and informative description, which does not always come to the same thing). When the relativity of classification systems (or hypothesis-testing methodologies) to their underlying pragmatic goals is left unstated, the result can be an uncritically narrow view of what should count as science, and also an uncritical belief that science, thus narrowly conceived, has a monopoly on approaches to discovering valid empirical truth.
These points may admittedly ring more true for (say) clinical psychology than they do for (say) physics. Even in physics, of course, there is debate about the value of theories that may place a higher premium on (e.g.) coherence than on testability, and which win coherence only at the price of countenancing entities and events that can never be observed. String theory comes to mind. But the basic categories into which physical stuff gets sorted are relatively low on the controversy scale, by comparison with the criteria according to which social "science" sorts human beings. DSM IV comes to mind.
Posted by: Peter Nordberg | Nov 24, 2006 3:58:33 PM
