Pursuant to an IM with Pascal, and subsequent email exchanges:

Science consists of two basic things: Facts, and guesses.

Sometimes, the facts change, sometimes what is accepted as scientific fact turns out not to be. But, for the most part, there are some hard facts that can be counted on to stay facts for a long time.

Temperature is one of them. temperature can be affected somewhat by atmospheric conditions, distance above sea level, etc. but for the most part, reading a mercuroid thermometer is a good indication of what the actual temperature is.

Now, let’s say, you want to know the temperature of a room. You can’t get a thermometer in the room, but you know the room contains a relatively active cat, a sedentary human, a steam register which is on 22% of the time, and a window which is open four inches, twenty inches wide, and it’s 34 degrees out the window.

You can just put a thermometer in a similarly sized room, but this isn’t very accurate. You can put the thermometer in the room with a window and a radiator, but that is only a bit more accurate. You can add a moderately active cat and a sedentary human, but that only narrows the margin for error. You can do the absolute best to duplicate the conditions of the room, and measure your mocked up room, but the fact is, there are so many things you don’t know.

Is the subject room insulated more or less than your room? Is the human in that room larger or smaller than yours? is there more or less skin? does the human in the subject room have a disease that makes him colder? warmer? does the radiator transfer heat efficiently, or is it covered in layer after layer of insulating paint? Is the window in the subject room facing into the wind most of the time, or away from it, or at some angle to it?

Even the process of measurement itself can affect the outcome. Like Schrodinger’s cat- if you open the door to the room to check the temperature you introduce potential variation caused by the opening of the door, your own body heat, etc.

The bottomline is, you can only guess. You can narrow the results of your guess, but you can still only guess. Science gives us the ability to generate scenarios that have less and less margin for error, and math allows us to calculate the amount of possible error.

That margin for error is the issue at work here. The past is a room we cannot enter. We can’t go there with a thermometer, so all we can do is measure the things we CAN measure, and compare them to other things we’ve measured. Frinstance: Look at how much CO is in the air now. Look at how much CO is in an air bubble found in an ice floe thousands of years old. That’s one way of guessing. Of course, we don’t know if that correlation is meaningful; it could well be that the CO levels are affected by different things at different times. So it’s a guess. Still, just a guess. Maybe a semi-scientific wild ass guess, but a guess nonetheless.

We cannot base global warming hysteria on guesses. If the planet is warming, we have to prove it by science, and then we have to prove that the warming is a bad thing. I’m not sure either is possible right now, but i know this: I know more than Algore about the scientific process.