Monday, February 25, 2008

Personal philosophy, part 3

On certainty:

Being “certain” about things is a concept far too many people just take for granted. Certainty is actually a complex and difficult subject, as it affects basically everything we do. Many people and organizations, to avoid discussing this complexity (since average people hate complex stuff), spend a comical amount of time trying to think up new language for sounding ever more certain, most of which ends up sounding at least a little contrived to most people.

The reality is that, for humans, there is no absolute certainty. Even in the most aware and intelligent of people, our perceptions and reasoning will always be at least slightly flawed. It’s just a fact of life. This leaves us with little choice but to figure out a way to measure and understand the, possibly small, but ever-present uncertainty in things.

Obviously, I’m interested both in finding ways to be as certain as possible about things and in figuring out when I am “certain enough”. I always start such thinking by considering how science deals with the issue of certainty. Scientists spend a great deal of time measuring things and doing calculations on those measurements. Since all of these measurements and calculations are done on numbers, “certainty” must also be “measured” as a number. Every measurement and calculation must have a value showing its certainty, or, actually, usually its uncertainty, since this is generally a smaller number and is really what the scientist is more interested in. I’ve been surprised to find that determining the uncertainty in measurements with most tools is left to the person taking the measurement; there may not always be a standard or system. There are a few guidelines, and the uncertainty for some measuring tools is well defined. The method for figuring out uncertainty in calculations, however, is very exact. Somehow, someone even came up with a set of formulas for finding uncertainty in calculations at some point.

So, how does this help me? It doesn’t, mostly. Daily decisions and personal philosophies generally can’t be condensed to numerical form, and you obviously can’t have a number that describes how certain a non-numerical idea is. I have gotten far enough in my pondering on this to conclude that the general idea behind scientific uncertainty (or at least the general idea as I see it; I’ve never heard anyone else describe it this way), may still be useful in other areas. So, I see general uncertainty like this: suppose you have a graph of the equation 1/x. It looks like a curved line going from nearly vertical at the y-axis to nearly horizontal at the x-axis. It never actually touches either axis, though, which is the key. Let’s now say that “x” is defined as “stuff I don’t know”, and that “y” is defined as “my level of understanding of stuff”. If “stuff I don’t know” is very big, “my level of understanding” gets very small, close to 0. Keep in mind, though, that it never actually equals 0. If “stuff I don’t know” gets very small, “my level of understanding” gets very big. “stuff I don’t know” can never actually be 0, though, so “my level of understanding” can never be infinite. That makes sense; we all know nothing in real life is infinite. The entire point of, well, everything we do, really, is to make “stuff we don’t know” smaller and “our level of understanding” bigger. We can keep eliminating little things we don’t know and understanding more and more, but our understanding is never infinite.

This is good enough for science because scientific measurements and calculations can be compared with previous measurements and calculations to determine whether they are more or less certain. Being more certain is good, being less certain is bad. There’s no need to talk about being “certain enough”. In real life, though, each experience is different, at least in some subtle way. We can and do gain experience and think about what is more likely to happen in a certain kind of situation, and we can act accordingly, but life will occasionally defy our expectations. Even subtle differences can have huge impacts. The bottom line, which I imagine most of you could have figured out without all the deep pondering, is that we largely make up our own meaning of “certain enough” as we go. And, for now, that will have to be good enough.

4 comments:

Unknown said...

Are you certain about uncertainty?

One For Logic said...

Why yes, yes I am.

Unknown said...

So wouldn't you be disproving the whole point about the permanent existing factor of uncertainty by being certain?

One For Logic said...

lol. I was joking. You make an interesting point, even uncertainty is uncertain.