You don't know what you want!

Sometimes discussion on other people's blog get so big, it is time to move at least part to another location. Neuroskeptic had a great post about the trap of extreme scientism, which combines the ideas that A) science can solve all problems and that B) one should not act except based on scientific evidence. The post is "My Breakfast With Scientism", and it starts with a man trying to determine which of two cereals he wants to eat. He realizes he needs some scientific evidence to determine which he should eat.... and things get much, much worse from there. Because an over-worship of science involves significant skepticism of normally routine claims and decisions, you end up in a very similar place to Descartes's over-worship of rationalism. You know:

Hey Descartes! You claim you are doubting everything.... but how can you be sure? }:- P

The comments on Neuro's post got into some really interesting ground that seems more appropriate for this blog... so I'm moving it over here. In particular, a commenter named "DS" asserted that "what he wants" is a scientifically admissible fact of the highest caliber. It isn't. But explaining why is can be difficult....



DS began by asserting:
"<i>I ask myself: What do I feel like eating for breakfast? And what ever that is it is a fact.</i>"



Indeed, what we wants is a fact - because it is what it is - so that seems pretty hard to argue with. But I want to give DS the benefit of the doubt that he was asserting something that wouldn't be true by definition. I believe he was asserting "Whatever I think I feel like eating for breakfast, it is a fact that I want that for breakfast."That is, I think he is asserting that the answer he gives us (or at least the answer he gives himself) is a factual representation of his desire.

Such would be a common claim. It derives most strongly from a Cartesian world view. According to Descartes (or at least the parts of his writing that people still read and discuss), we have infallible knowledge of our own minds. However, this odd assumption has been challenged many times, in many ways. The rejection of this assumption is a hallmark of the pragmatist philosophers, who (at least in the old versions I am most familiar with) treat your claims about yourself as hypotheses on equal footing with your hypotheses about anything else.

I don't want to get into the nitty gritty details of how different philosophers have made this argument, but it is important to agree at this point that some people don't know what they want. We don't even need Bugs Bunny to prove it (but you can use Bugs if you want). Mundane cases abound, such as this pretty routine occurrence in which my daughter says: "I want the soup" "Honey, I tasted the soup, and you don't want it"... <they take a taste>... "I don't want the soup, can I have something else." || Or, if you think kids should be granted an exception, we can turn to a standard emerging-adult event. A student walks into my office: "I want to be a forensic psychologist" "Do you read books about forensic psychology in your spare time?" "No" "Do you know anything about the profession, or what education you will need?" "No" "Are you good as science?" "No" "Well, then you don't want to be a forensic psychologist" "Oh, yes I do" || The same potential problem holds when someone goes to get breakfast. What you think you feel like eating for breakfast is not an infallible indicator of what you want for breakfast.

If I have to ask you what you want for breakfast, and bring it to you, the problem is compounded. I'm pretty sure we would all agree that your verbal report is not a scientifically admissible "fact" about your wants.... it is just a verbal report. If we add in a bunch of other assumptions --- you "know the language", are not trying to deceive me, etc. --- then you are probably very accurate in verbally reporting your desires. But we need those assumptions, and even then we would only expect you to be highly accurate, not infallible.


If I can at least get you to admit that your verbal behavior is a fallible indicator, then we are on the right track. When you ask the little voice in your head a question, and the little voice answers you, we have all the same problems. If you wake up in the morning and say to your self, "What do I want for breakfast?" and then you say to yourself "Cornflakes, yes, I want cornflakes", you are stuck in a one-person version of the verbal-report problem. The voice in your head has made an assertion, but you have no guarantee it is accurate.

When the little voice answers you, you have, at best, a reasonable hypothesis as to what you want. Nothing more. Maybe less. What you <i>actually</i> want is an empirical question that can be investigated scientifically. Hence, someone taking a stance of <i>extreme</i> scientism <i>would</i> indeed be stuck in the conundrum Neuro lays out.