Saturday, November 12, 2005

Science/Religion - ID musings


I came across an argument for ID that I haven't heard before, on Scott Adams's blog (he of Dilbert fame) :

Here’s where it gets interesting. The Intelligent Design people allege that some experts within each narrow field are NOT convinced that the evidence within their specialty is a slam-dunk support of Darwin. Each branch of science, they say, has pro-Darwinists who acknowledge that while they assume the other branches of science have more solid evidence for Darwinism, their own branch is lacking in that high level of certainty. In other words, the scientists are in a weird peer pressure, herd mentality loop where they think that the other guy must have the “good stuff.”

Is that possible? I have no way of knowing.

But let me give you a little analogy. One time in my corporate career I was assigned to lead a project to build a 10 million dollar technology laboratory. The project was based on the fact that “hundreds of our customers” wanted a place to test our technology before buying our products. I interviewed several managers who told me the same thing. Months into the project, I discovered that there was in fact only one customer who had once asked for that service, and he had been satisfied with another solution. The story of that one customer had been told and retold until everyone believed that someone else had direct knowledge of the hundreds of customers in need. If you guessed that we immediately stopped the project, you’ve never worked in a big company. We just changed our “reasons” and continued until funding got cut for unrelated budget reasons.


(While you may be thinking that I should probably know better than to give my uninformed opinion on this highly complex matter, keep in mind this blog's slogan.)


Here is where I think Scott's analogy (and argument) goes awry. There are two differences between a business project and scientific research, that immediately spring to mind.


Firstly, within a company, there tends to be a great deal of secrecy when it comes to information sharing. Departments tend to limit information only to the relevant people to reduce the risk of it being stolen by rivals or even rival departments within the same company (after all, knowledge is power). On the other hand, academia is completely the opposite: if experiments are not verifiable, then the results are not recognised. Consequently, all the relevant information is available for everybody to analyse for themselves. Scott's scenario would have never occured, had the managers involved hard data on the customers.

Secondly, a business project has a pretty strict grouping (e.g. these 10 people work on the hardware, these 10 people work in sales, etc), and generally as a consequence, specialisation is greater: one group probably doesn't know the details of what the other group is doing. Academia, on the other hand, has no such rigid boundaries. You have plenty of biologists that are involved with chemistry and mathematics, you have plenty of mathematicians and chemists that are interested in biological problems. So you have many cross-discipline people that understand the problems and arguments of both domains, can critically analyse the weaknesses and strengths of each. Considering that they have access to the hard "evidence" and information from both areas, these people should be able to red-flag any "herd mentality" and bridge the misconceptions arising from not being a specialist in one field but not the other.


Besides, I see no problems with starting out with the assumptions that Darwin is right. How a scientist proceeds is that he/she makes some assumptions, builds the model from those assumptions, predicts theoretical results from the model and compares these predicted results with real world results. If the results match up, then the scientist gains confidence, and puts more "trust" in the assumptions. Alternatively, if the results seem off, the scientist reexamines the assumptions, comes up with some new ones, and runs the whole process again. Nobody forces those who are unhappy with Darwin to use his theories; if they can come up with something that yields more accurate predictions, great!


I'd be interested to see more evidence from "pro-Darwinists who acknowledge that while they assume the other branches of science have more solid evidence for Darwinism, their own branch is lacking in that high level of certainty". While they probably wouldn't be able to shatter Darwin's theories completely (those theories have allowed many different research fields to generate very accurate models), they would probably provide modifications and adjustments which "feed back" into the core Darwin stuff: that way everybody is happy.

No comments: