Yesterday I made the point that Jim Hansen’s latest threat of doom was a conditional one. I also said that there was nothing wrong with this. And there isn’t. Nothing wrong with the structure of a contingent—which is to say conditional—prediction. I hinted, in a way, that better predictions were not so conditional. This is true in a sense which I did not include in my essay, because in fact, all predictions are contingent. And so is everything else.
To clarify, but briefly.
Any prediction is a statement that, given some set of information or evidence or revelation or knowledge, that “Y will happen” or that “There is a P% probability Y will happen.” You cannot just step onto to the street and announce, “The end is nigh” without there being some shared, or assumed shared, conditional set of knowledge which your listeners hold.
It also goes without saying that your audience must understand what “end” and “nigh” mean. If your prediction fails, there are more way to escape responsibility by disputing what you really meant by “end” than there are Democrats in San Francisco.
The knowledge of word (or mathematical or whatever) definitions are thus also part of the conditional knowledge which is at least tacitly assumed or sometimes explicitly given. For example, Hansen said (I’m paraphrasing) “If Canada sells it oil and the USA doesn’t do something, the Four Horseman begin their ride.” The conditionality is explicit in part, and stated boldly—in part.
The parts that are missing are where the trouble arises. Firstly, Hansen does not make it clear what “USA doing something” means. What exactly is “something”? This kind of seemingly precise vagueness is the trade secret of five-dollar psychics who pass off cold readings as genuine paranormal knowledge. Hansen’s prediction is in the form of, “I see a U. Maybe with an S or an A. Does that mean anything to you?” He lets his listeners fill in the meaning, which invariably is done in the most charitable way.
Secondly, Hansen’s unleashing of the apocalypse (this is his word) is blurry. He mentions food prices rising to “unprecedented” levels. What does that mean exactly? He also says that we’ll see more bad weather. Well, given inflation and the assumption that we’ve seen bad weather in the past, neither of these events can be said to be rare. Lack of specificity plagues his predictions of plague.
Now I suggested yesterday, in poor, even rotten, language that Hansen should have issued his prognostication in unconditional form. Since this is a logical impossibility, I was as wrong as can be. What I meant, but was too lazy or stupid to clarify, is that he should make his conditions explicit, testable, verifiable, measurable, real.
As should anybody who makes a prediction. This includes people who try to pass off failed predictions as “scenarios.” A scenario is a prediction like any other, but usually issued as a cluster. “If A happens, then Y_A will happen”, “If B happens, then Y_B will happen”, and so on. Well, in the end, if it clear what time horizon is meant, we look to see which of Y_A, Y_B, … actually occurred. Suppose Y_R did. We then look to see if R happened, too. If so, then we have a good prediction. If instead B happened, the forecast is a bust.
In other words, you cannot issue any statement without there being something to back it up. This is the conditionality. Regular readers met this earlier when we spoke of statements of knowledge and of probability. There are no unconditional statements of either, except in one sense, to be explained in a moment.
Above I gave a conditional statement of probability “There is a P% probability Y will happen” but left off the conditions though said they were always there. That means the true statement is really “Given X, there is a P% probability Y will happen.” If you doubt this, I challenge you to discover even one instance of a probability which is not conditional, which does not require a set of evidence or knowledge for its definition. (JH, this one’s for you.)
And the same with knowledge. As we’ve used a hundred times, we know it is true that “Socrates is mortal” only because we accepted the conditions that “All men are mortal. And Socrates is a man.” And because we knew that arguments like this lead to true conclusions. (We also knew what the words meant.)
But how did we know about arguments like this? Well, we just knew. There are some basic, or base truths, which we just know are true. I have earlier said that even these are conditional, and this is so. We say we know that “If x = y then y = x” is true conditional on our intuition.
Our friend Luis objected to the use of the word “intuition.” I can see his point. Let us instead say that all men have these basic truths written in their hearts.