
                         Roanoke Times
                 Copyright (c) 1995, Landmark Communications, Inc.

DATE: FRIDAY, May 26, 1995                   TAG: 9505260037
SECTION: EDITORIAL                    PAGE: A-18   EDITION: METRO 
SOURCE: DAVID A. DEWOLF
DATELINE:                                 LENGTH: Medium

METHINKS THE IDEA ISN'T SO RIDICULOUS

THOMAS Foley (May 20 letter to the editor, ``Randomness leads to violence'') ridicules Morton Nadler (May 7 commentary, ``Such a God as this is no comfort'') on the grounds of faulty logic. He also finds Nadler's philosophy as ``empty'' as his logic, and suggests Nadler check his dictionary on concepts such as randomness, laws and ethics.

Foley ridicules the idea that random events can lead to organized systems. That idea isn't ridiculous, not from a scientific standpoint.

Let me paraphrase an eye-opening example cited by Nobel laureate and DNA-structure discoverer Francis Crick in his book ``What Mad Pursuit.'' He cites Richard Dawkins' book (``The Blind Watchmaker'') that discusses a simulation by two computer processes of random construction of the Shakespearean sentence: ``Methinks it is like a weasel.''

In the first single-step selection process, each of 28 letters and spaces in the sentence is generated sequentially and randomly. You would have a one in 10-to the-power-40 chance of creating this sentence this way. The time required to do so far exceeds the age of the universe.

In the second cumulative-selection process, a small number of 28-symbol copies of a random sequence of 28 letter-space symbols are generated, each with slight deviations (``mutations''). The computer program accepts only one of these that most resembles the target sentence above, however slightly. The process then repeats itself with the slightly ``improved'' version. The surprising but essential element in this is that it takes considerably fewer than 100 steps (on average) to reproduce accurately the sentence in question.

While this procedure deviates somewhat from nature, in that nature doesn't know where it's going, Dawkins' refined versions of this experiment should convince an unbiased mind that natural selection of essentially random mutations can lead to highly organized entities in biologically possible time limits.

With respect to ethics, it's not unreasonable to think that universal condemnation of murder and theft helps maintain law, order and stability in a human society. All Ten Commandments seem to me to fall within that category. If you consider ethical consistency as a result of most societies' conforming to a pattern that ensures the greatest likelihood of survival, then the development of ethics, too, could fall in the realm of the above cumulative-selection process of evolution.

Foley would benefit by consulting some basic science texts before admonishing others to check their dictionaries.

David A. deWolf, of Blacksburg, is a professor of electrical engineering at Virginia Tech.

by CNB