HOME | POLITICS | SPORTS | LIFE | SCI/TECH | OPEDS | HELPFUL TIPS
|
Sept. 23, 2004 There are many important examples, in political arguments, and in environmental laws and lawsuits, where two things happen at the same time, and it is assumed that one of them caused the other. One example is the fact that several decades ago, frogs in the northern states of the U.S. began to disappear, just at the same time that tiny amounts of "PCB" (polychlorinated biphenyl) began to be detectable in the Great Lakes and other big waterways. Does that prove that the PCB caused the frogs to die out? Not really, because there might be a third factor that caused both, like changes in wind patterns that blew PCB from industrial areas onto the Great Lakes watershed, and maybe the new winds also interfered somehow with frog reproduction. Or maybe there was some different additional factor, or possibly even just "random variation." Frog populations are known to increase and decrease from time to time, seemingly at random. (Of course, a well known case of random variation is that, if you toss a coin ten times, you might very well get 7 or 8 heads, even though five would be expected to occur on the average. There would be no known "cause" of the variability of your results --- they are just random.) Judges, juries, and legislators often don't seem to understand these points, which can have grave consequences in the ways our laws are made and enforced. Companies have gone bankrupt with thousands of employees laid off, and doctors and engineers have been successfully sued for millions of dollars, when lack of understanding among legal people allowed that to happen. (The author has been an expert witness and a consultant in a few such cases.) Here is a clearcut example from the author's industrial experience, working in the old Western Electric division of AT&T, before the Richmond factory became part of Lucent Technologies. One fine April day, the author and his very capable technician were installing a new chemical process in Virginia. The process worked during the first day, but then while driving from our motel to the factory the next morning, the technician noticed myriads of beautiful forsythia bushes beginning to bloom. For the rest of that day, the process utterly failed to work. Soon after arriving home in disgrace up north at our New Jersey research lab, the process did work again as it previously did there. However, a few days after that, the forsythia bushes blossomed up north where we then were, as the summer weather crept northward. The process utterly failed once again. We knew that very small amounts of some kinds of "dirt" could kill our sensitive process. Therefore we seriously wondered if pollen from the forsythia blossoms had sabotaged us. There was a lot of money involved, and our upper management was getting madder by the minute! It turned out, after much agonizing, that a third factor had caused both the forsythia blooms and the process crash: it was a sharp rise in the humidity from nighttime rains in April. Extreme measures to remove moisture solved the problem in the lab and later in the factory, and the process was run successfully for many years afterwards. For anyone wanting to read five more stories like this (which sometimes make the author look pretty bad, at least temporarily), you can click on this link: The Green Monster It just goes to show you, maybe the frogs are dying, but things could be worse, although most such things don't actually happen often. ------------ About the author: Dan Shanefield is a retired engineering prof, who worked at Bell Labs and then at Rutgers University. He wrote the book "Industrial Electronics for Engineers, Chemists, and Technicians". Visit his website or email Dan Shanefield: shanefield@ieee.org Tell a friend about this site! ------------ |
||||||
|
|
|||||||
|
