So seeing many white swans makes you less prepared for black swans than someone who has seen 0 swans?
Yes, but it is even more complex than that as it is pointing to the communication of private experiences. Let’s say that you see the swans as one category but someone tells you that there are some swans which he calls ‘black’ and are different. You look closely but you only see white swans. Then he tells you that this is because your perception is not trained correctly. You ask for evidence and he says that he doesn’t have any but he can tell you the steps for seeing the difference.
The steps have to do with a series of exercises that you have to do every day for a year. You then have to look at the swans at a certain time of the day and from a certain angle so that the light is just right. You look at the person and wonder if they are crazy. They, on the other hand are in a position where they do not know how to convince you. You rationally decide that the evidence is just not enough to justify the time investment.
After a number of years that person comes up with a way to prove that the result is genuine and also demonstrates how the difference has consequences in the way the swans are organized. You suddenly have enough rational evidence to try! You do and you get the expected results.
The knowledge was available before the evidence crossed the threshold of rational justification.
I guess with swans you can just say “go look at swans in africa” which gives a recipe for public experience that would reproduce the category-boundary.
It is the case with seaguls that they are mostly white but males and female have different ultraviolet patterns. Here someone who doesn’t see ultraviolet can easily tell the classes apart but someone who sees only 3 colors will be nearly impossible. Then if throught some kind of (convulted) training you could make your eye see ultraviolet (people with and without all the natural optics respond slightly differently to ultraviolet light sources so there is a theorethical chance some kind of extreme alteration of the eyes could yield it to clearly recognisable levels).
Now ultraviolet cameras can be produces and those pretty much produce public experiences (ie the output can be read by a 3 color person too). Now I am wondering whether the difference between “private sensors” and constructed instruments is merely that with constructed instruments we do have a theory how they work but with “black box sensors” we might only know how to (re)produce them but don’t know how they actually work. However it would seem that sentences like “this and this kind of machine will classify X into two distinct groups Y and Z” would be interesting challenges to your theory of experiment setting and would warrant research. That is any kind of theory that doesn’t believe “in the training” would have to claim something other on what the classification would be (that all X would be marked Y, that the groups would not be distinct, that the classifier would inconsistently label the same X Y one time and Z the next time). But I guess those are only of indrect interest if the direct interest is whether groups Y and Z can be established at all.
Hehe, I didn’t mean it that literal just trying to get the idea across :)
Nevertheless, your analysis is correct for the case where alternative ways of confirmation are available. There is of course the possibility that at the current stage of technological development the knowledge is only accessible through experience like in my lucid dreaming example in the original post.
Yes, but it is even more complex than that as it is pointing to the communication of private experiences. Let’s say that you see the swans as one category but someone tells you that there are some swans which he calls ‘black’ and are different. You look closely but you only see white swans. Then he tells you that this is because your perception is not trained correctly. You ask for evidence and he says that he doesn’t have any but he can tell you the steps for seeing the difference.
The steps have to do with a series of exercises that you have to do every day for a year. You then have to look at the swans at a certain time of the day and from a certain angle so that the light is just right. You look at the person and wonder if they are crazy. They, on the other hand are in a position where they do not know how to convince you. You rationally decide that the evidence is just not enough to justify the time investment.
After a number of years that person comes up with a way to prove that the result is genuine and also demonstrates how the difference has consequences in the way the swans are organized. You suddenly have enough rational evidence to try! You do and you get the expected results.
The knowledge was available before the evidence crossed the threshold of rational justification.
I guess with swans you can just say “go look at swans in africa” which gives a recipe for public experience that would reproduce the category-boundary.
It is the case with seaguls that they are mostly white but males and female have different ultraviolet patterns. Here someone who doesn’t see ultraviolet can easily tell the classes apart but someone who sees only 3 colors will be nearly impossible. Then if throught some kind of (convulted) training you could make your eye see ultraviolet (people with and without all the natural optics respond slightly differently to ultraviolet light sources so there is a theorethical chance some kind of extreme alteration of the eyes could yield it to clearly recognisable levels).
Now ultraviolet cameras can be produces and those pretty much produce public experiences (ie the output can be read by a 3 color person too). Now I am wondering whether the difference between “private sensors” and constructed instruments is merely that with constructed instruments we do have a theory how they work but with “black box sensors” we might only know how to (re)produce them but don’t know how they actually work. However it would seem that sentences like “this and this kind of machine will classify X into two distinct groups Y and Z” would be interesting challenges to your theory of experiment setting and would warrant research. That is any kind of theory that doesn’t believe “in the training” would have to claim something other on what the classification would be (that all X would be marked Y, that the groups would not be distinct, that the classifier would inconsistently label the same X Y one time and Z the next time). But I guess those are only of indrect interest if the direct interest is whether groups Y and Z can be established at all.
Hehe, I didn’t mean it that literal just trying to get the idea across :)
Nevertheless, your analysis is correct for the case where alternative ways of confirmation are available. There is of course the possibility that at the current stage of technological development the knowledge is only accessible through experience like in my lucid dreaming example in the original post.