Posts
The neo-Darwinian theory of evolution claims that evolution of living organisms happens as natural selection acts on random genetic mutations. But what does it mean to say that the mutations that form the basis for Darwinian evolution are "random"?
Some proponents of Intelligent Design reject neo-Darwinian evolution because of the common belief that saying something is random is equivalent to saying it is unplanned and unguided. This belief, I suggest, is mistaken. To say some process is random says nothing about whether it is planned or not. In the context of scientific mathematics, randomness is a technical term that neither includes nor implies such common understandings as "haphazard", "aimless", or "lacking plan or purpose". These form no part of the scientific or statistical definition of "random".
Randomness is a mathematical property that means one (or more, depending on the situation) of the following: "uncorrelated", "unbiased", "unpredictable". This is how the word is used in statistics, mathematics, and science, including scientific discussions of biological evolution.
As a statistician, I deal with randomness all the time. In sample surveys, for example, randomness in selection of respondents is a highly desirable—indeed, essential—property that we take great pains to realise. Our surveys are painstakingly planned and implemented with the highest degree of technical proficiency we can muster. Randomness in my line of work doesn’t just happen: we have to prepare carefully in order to achieve it. Randomness, one might say, is a question of design. (I’ll return to that later.)
In this connection, Dr Stephen M. Barr of the University of Delaware has written an excellent article entitled "The Design of Evolution" in a recent issue of First Things. The complete text of the article is available here, and I commend it to everyone interested in the evolution-creation controversy. Dr Barr offers two analogies to illustrate statistical randomness:
My children like to observe the license plates of the cars that pass us on the highway, to see which states they are from. The sequence of states exhibits a degree of randomness: a car from Kentucky, then New Jersey, then Florida, and so on—because the cars are uncorrelated: Knowing where one car comes from tells us nothing about where the next one comes from. And yet, each car comes to that place at that time for a reason. Each trip is planned, each guided by some map and schedule. Each driver’s trip fits into the story of his life in some intelligible way, though the story of these drivers’ lives are not usually closely correlated with the other drivers’ lives.
Or consider this analogy. Prose, unlike a sonnet, has lines with final syllables that do not rhyme. The sequence those syllables form will therefore exhibit randomness. But this does not mean a prose work is "unguided" or "unplanned". True enough, the writer did not select the words with an eye to rhyming them, imposing on them that particular kind of correlation. But the words are still chosen. So God, though he planned His work with infinite care, may not have chosen to impose certain kinds of correlations on certain kinds of events, and the motions of the different molecules in a gas, for example, may exhibit no statistically verifiable correlation.
This leads to the conclusion that statistical randomness is by no means incompatible with God’s governance of His creation through divine Providence. To God, nothing is unplanned. God may simply have chosen to allow events to proceed through natural processes that appear to us to be uncorrelated or unpredictable.
To return to my professional experience again: a common technique used to contact respondents in surveys conducted by telephone is random digit dialing. How are the telephone numbers generated? I hope no one is surprised when I say that we do not instruct our interviewers to close their eyes and hit a few buttons on a touch-tone phone. Rather, a computer program is written to generate lists of random telephone numbers, which are handed to the interviewers with instructions to dial them sequentially until a genuine number is dialed. We have found that it takes, on average, about ten randomly generated telephone numbers to find one useable number for survey purposes.
This brings up another interesting perspective on this randomness business. Computers, even hand calculators, can generate random numbers. But computers can only work on the basis of instructions, i.e., they need an algorithm. Computers are exceedingly deterministic: they do exactly what they’re told—nothing more and nothing less. Given knowledge of the algorithm, the entire list of computer-generated numbers is perfectly predictable. So, strictly speaking, computer-generated "random" numbers are not random. It’s just that the algorithm is so inventive and so extremely complicated that the result is close enough to random that the numbers are uncorrelated, as far as can be detected with the computing power available. To be precise, some mathematicians call such lists of numbers "pseudo-random".
The professor of the first statistics course I ever took once said something like, "With enough data points and enough computing power, everything can be predicted". He was referring specifically to economic phenomena, but what if those random genetic mutations that form the building blocks for evolution aren’t really random, but biologists just don’t yet have enough data points and computing power to tell? What if God has used an extremely complex and intricate algorithm to generate "random" mutations? If so, then maybe some day, if the Lord delays his return, when more data observations have been gathered, and computers are millions of times more powerful than they are now, some biologists will announce that they have discovered hither-to unknown correlations in genetic mutations. They may be able to reveal a fantastically complicated algorithm to account for such correlation between mutations.
If that ever happens, Einstein will once again have been proven right when he said, "God does not play dice with the universe".
I realize that quantum physics, about which I know very little, has something to say about this. However, I've reached the limits of my scientific knowledge.
This cartoon by Matt appeared in the