Friday, November 15, 2013

Are Things Really Random?

How can an algorithmic process like computer programming actually produce a random number? I was avocationally interested in understanding random-number generators. I looked into it.

Turns out, a computer program can't actually generate random numbers. So-called random number generators do not, in fact, generate random numbers.

It's more complicated than this, but the principle is easy: Choose a string of numbers in which you can't find a pattern, like, say, a particular chunk of the irrational square root of 2, and then use that sequence to determine the outcome you want to generate "randomly."

Why bother to point this out?

Well, one of the core assumptions of our modern minds, often unquestioned, is that any number of phenomena occur randomly. Without the assumption of randomness, statistics, a cornerstone of many scientific methods, is undermined.

It seems to me that there are three objections to the assumption of randomness.

1. From physics: Brilliant physicists, like Stephen Hawking, tell us, "Everything is determined. But [because of things like the uncertainty principle] we can't know what has been determined." What appears random to us, then, is actually deeply embedded in the motion of the universe and could not have happened otherwise. So randomness is simply an assumption that leads to an approximation of what we cannot otherwise fathom.

2. From statistics: Even within this framework, it's clear that some phenomena appear random to us (like coin flips) and others do not (like my decision not to have seconds on dessert tonight). But any statistician can tell us that it's rarely, if ever, clear whether or not any particular phenomenon is random. We can analyze results with a degree of confidence, but confidence is simply not certainty.

3. From history: Ancient persons looked at the sky and saw meaning in the arrangement and movement of the stars. To them, many, many phenomena that we find random, they found meaningful. Today, skeptics have "proved" that astrology is false. But, really, you can't prove such a thing, you can simply call it into question. Believers believe, and non-believers disbelieve. True skeptics remain... skeptical, unconvinced, willing to be proved wrong. Or right.

I choose astrology because I am particularly indifferent to it. I have not found evidence of its truth (or falsity) in my life, and I've made it so far as a happy person thinking about lots of other things. If astrology contains truths, I'm blissfully unaware of them. And, if it's false, it's of no consequence to me.

The world allows for more than two positions, of course, and I believe most of us live somewhere between cold, statistical models and a feverish devotion to every "meaningful" omen.

My points, in the end, are not particularly large, but they may have large implications. If the world is less random than many of us commonly assume, we can learn by being open to new patterns of meaning of which we were previously ignorant. It's good to be clear about what we know, what we don't know, and what we assume but may be wrong about.

Last, with the assumption of randomness comes the apparent discovery of meaninglessness. If we value meaning--or, at least, the search for meaning (which is a value and a meaning in itself)--then we owe it to ourselves to chafe against the tyranny of randomness, not to believe in it when we aren't called to, and to keep our minds appropriately open.

1 comment:

Anonymous said...

You would very much enjoy this book: