MoreOn comments on Rationalization - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (25)
Try answering this without any rationalization:
In my middle school science lab, a thermometer showed me that water boiled at 99.5 degrees C and not 100. Why?
I suspect you have a point that I'm missing.
My take is: either the reading was wrong (experimental error of some kind), or it wasn't wrong. If it wasn't wrong, then your water was boiling at a 99.5 degrees. There are a number of plausible explanations for the latter; the one that I assign the highest prior to is that you were at an elevation higher than sea level.
So, my answer is in the form of a probability distribution. Give me more evidence, and I will refine it, or demand and answer now, and I will tell you "altitude", my current most plausible candidate (experimental error is my second candidate, first with how (where in the water) you measured, then with the quality of the thermometer. After that trails things like impurities in the water).
What altitude were you at?
My experience leads me to assume that the thermometer was mismarked. My high school chemistry teacher drilled into us that the thermometers we had were all precise, but of varying accuracy. A thermometer might say that water boils at 99.5 C, but if it did, it would also say that it froze at -0.5 C. Again, there are conditions that actually change the temperature at which water boils, so it's possible you were at a lower atmospheric pressure or that the water was contaminated. But, given that we have a grand total of one data point, I can't narrow it down to a single answer.
What elevation was your school at?
You've missed a key point, which is that rationalization refers to a process in which one of many possible hypothesis is arbitrarily selected, which the rationalizer then attempts to support using a fabricated argument. In your query, you are asking that a piece of data be explained. In the first case, one filters the evidence, rejecting any data that too strongly opposes a pre-selected hypothesis. In the second case, one generates a space of hypothesis that all fit the data, and selects the most likely one as a guess. The difference is between choosing data to fit a hypothesis, and finding a hypothesis that best fits the data. Rationalization is pointing to a blank spot on your map and saying, "There must be a lake somewhere around there, because there aren't any other lakes nearby," while ignoring the fact that it's hot and there's sand everywhere.