Woozle/drafts/rationality/v1
This is an early draft of the Rationality essay. I don't consider it valid overall, though there are valid bits of it. I'm mainly reposting it here so I have an easily-accessible record of where I was with this argument conceptually when I first started setting it down on 2010-08-07. --Woozle 19:19, 23 August 2010 (EDT)
(Disclaimer: to whatever extent it is necessary to do so, I recognize my own inadequacy and unworthiness to be addressing this topic, which has been so much more knowledgeably and intelligently explored by others. I am not attempting to criticize, diminish, or usurp their work, nor do I claim to be building upon it. I recognize that I am not worthy to lick EY's boots, and I don't mean that sarcastically (hyperbolically at worst). What I am attempting to do is to fill a gap I perceive. If the gap has already been filled, then please point me at the work I have overlooked. Thank you.)
In an earlier discussion, I argued that someone was not being rational. They replied that I couldn't define rationality, and so the question was moot.
I went "dot dot dot", and had to go off and think about it for awhile. I mean, here we have an entire site dedicated to a particular concept, and yet we have no working guidelines by which we can determine whether that concept applies in a given case?
I've now thought about it enough that I have a tentative answer.
I propose that rationality is:
- (short) the act of using documented reasoning processes.
- (medium) the act of using only documented reasoning processes, so that they can be duplicated and refined or corrected.
- (long) is the act of using only documented reasoning processes to reach a decision so that the accuracy of one's premises and logic can be verified and/or corrected.
I also propose that this involves two key concepts (language scholars please suggest better words):
- Rationifying: the act of putting a reasoning process into a logically-analyzable form
- Ratifidelity: the act of heeding the conclusions of a rationified thought-process, rather than allowing other non-documented processes to override those conclusions
In order for an act to be rational, you have to have both of these -- i.e. the thought-process you used to arrive at your decision must be logically correct and you must make use of the conclusion at which the logic arrives.
"Make use of" can be as minimal as simply announcing that this is your decision. If you then go off and take a different action from the one reached by the process, we would say your announced decision was rational while your actions were not. Similarly, if you announce a decision without explaining the logic, then whatever degree of uncertainty exists about what your actual thought-processes were is also the degree of irrationality involved in the decision. It doesn't matter if you get the "right" answer if we can't replicate your thinking, because while different thinking processes might lead to the same answer in this situation, they might lead to very different conclusions in a slightly different situation. The process is the key, because it is the reputation of the process that is at stake, because only processes can be passed on to new decision-makers.
(Side note: if you regularly have, say, "hunches" in a certain area, and you have carefully observed these hunches and noted that they have a certain success rate, and it so happens that this success rate is higher than that achieved by any other process yet known, then ratifidelity would not be violated by following that hunch. You would be using methods of rationality to gain the best use of a non-rational input. This does, however, open up a whole grey area which I'm not going to write about just now.)
In the process of our discussion, my opponent had stated pretty clearly that he felt no obligation to help maintain civilization or assist others less able, and yet he seemed to expect me to find his argument persuasive.
I argued that this was irrational, because there was no rationified thought process which would argue that he should do this. I suggested to him that:
- If you want to persuade someone else of your point-of-view, you either have to convince them that your argument is in their personal best interest or that it is in some larger best interest.
- My opponent made it clear that his interests were entirely selfish.
A rational response to that argument would be to point out some logical argument which did argue for the actions he took, to admit that there was no reason for me to find his arguments persuasive, or to point out a logical flaw in my argument.
In the absence of any of those options, he had clearly failed both to rationify (to produce an argument of his own) or to achieve ratifidelity (failure to follow the logical conclusions in the available argument).
So... what am I overlooking here?