Thinking as a Hobby


Home
Get Email Updates
LINKS
JournalScan
Email Me

Admin Password

Remember Me

3478433 Curiosities served
Share on Facebook

Who Cares if Theories are Scientific?
Previous Entry :: Next Entry

Read/Post Comments (3)

As I said in the last post, I just finished watching Nova's Judgment Day, about the Dover, Pennsylvania trial in which Intelligent Design proponents tried to slip religious content into the school curriculum and soundly got their asses handed to them.

There were a couple of things I didn't like about the program, though. One had to do with the famous bacterial flagellum explanation. Michael Behe argued that the flagellum, an intricate motor, is irreducibly complex. That is, all of its parts are necessary for it to function properly, and so it could not have evolved incrementally. All of its parts would need to be created at one point in time. As a refutation of this, the program presented a simpler bacterial structure with a subset of the parts used in the flagellum, functioning as a type of syringe. While this is a useful example of how a simpler structure could be complexified and then co-opted for an entirely different function, it doesn't directly address the point of Behe's, which is that to function as a motor, the flagellum needs to have all its current parts. If they had shown a simpler structure, still used in locomotion, I think the example would have been much more compelling.

Kenneth Miller actually wore a mousetrap as a tie clip to make a similar point, removing parts of the mousetrap to show that it could still serve an intermediate function. But again, it doesn't work as a mousetrap.

A better example are eyes, which Dawkins discusses in The Blind Watchmaker at length. What good is half an eye? Or 10% of an eye? A hell of a lot better than having no eye, Dawkins answers, before laying out the overwhelming case for the evolution of eyes, showing the spectrum of structures from simple clumps of photoreceptive cells to concave structures to those with lenses to very complex eyes such as our own. At each step along the way, though, the structure still serves the same function: to process visual input from the environment. So it works as a much better refutation of irreducible complexity.

Another issue I was thinking about while I watched the program related to the topic of falsification I posted about recently. At one point in his testimony, the prosecutors got Behe to admit that under his criteria, astrology was a scientific theory. For this, Behe was ridiculed, and it was seen as a pivotal point in the trial and a heavy blow to the defense.

But while I have little sympathy for Behe, I do have some sympathy for the idea that labeling a hypothesis or theory "scientific" may not be all that useful. What does it mean for a hypothesis to be "scientific"? If it is simply that it is stated in a way that leads to falsifiability, that seems like a poor basis. As Kuhn pointed out, scientists rarely reject theories on the basis of anomalies. But if we don't use falsifiability to determine whether a hypothesis is scientific or not, then what criteria do we use? We could make the distinction on whether or not the hypothesis invokes supernatural causes or relies solely on the assumption of naturalistic causes, but while that might be a useful way to make the distinction, it might also carry us further afield, arguing about what makes a given cause naturalistic or not.

Instead of steering that way, I'd be more inclined to say the labeling of a hypothesis as "scientific" or not is probably not all that useful. Instead, I'd prefer to just talk about what makes a given hypothesis or theory good or bad.

I'd say three relevant criteria for determining the quality of a hypothesis are that it:

  • is parsimonious
  • explains current data
  • makes predictions regarding unknown data


Basically, does the hypothesis attempt to explain the phenomenon without invoking unnecessary complications? Does the existing evidence fit the hypothesis well? And does the hypothesis make predictions for which new data can be gathered to see how well it fits?

If we use these criteria, it doesn't really matter whether the hypothesis is "scientific" or not. All we care about is how efficient and effective it is. And on these counts, theories such as astrology and Intelligent Design are next to worthless, while theories such as General Relativity and evolution are extremely good.

ID is not parsimonious, in that the cause for complex biological features is a designer, whose qualities, intent, and complexity cannot be assessed and whose own existence must then be explained if the theory is to be reasonably complete. It does not explain the timeline of the fossil record, and cycles of speciation and extinction, and the very clear implication of an interrelated tree structure to the history of life rather than independent creation of discrete species. And it cannot make predictions, if the basis for new structure or species is the will of an undefined intelligent agent, unless ID can explain the will of such an agent. Is this agent going to make new species? Or is it done with creation? For ID to make predictions is to say that it knows the mind of god.

And so it goes for other crummy theories like astrology or numerology or cryptozoology, and with other shoddy ideas like conspiracy theories. I don't really care if they fall under the rubric of "scientific" or not. What I care about is whether or not they cleanly and effectively explain some aspect of the world and make predictions about some aspects we haven't observed yet.


Read/Post Comments (3)

Previous Entry :: Next Entry

Back to Top

Powered by JournalScape © 2001-2010 JournalScape.com. All rights reserved.
All content rights reserved by the author.
custsupport@journalscape.com