Edit: Now I see Sister_Y addressed my point in the very next paragraph, so this entire comment is a reading comprehension fail more than anything.
Necroing:
poke—my friend likes to explain this to his undergrads by asking them how they would verify that a thermometer is accurate (check it against another thermometer, but how do you know that one is accurate . . . etc.) until they figure out that thermometers are only “accurate” according to custom or consensus. Then he asks them how they know their eyes work. And their memories.
Some of them cry.
Go to the beach, light a fire, boil some water. Put the thermometer in the boiled water—does it show 100 degrees Celsius?
Still at sea level, put a cup of water in a fridge untill it starts freezing. Put the thermometer in the cup—does it show 0 degrees Celsius?
If yes to both, you have a working thermometer. This way, you don’t rely on the consensus of other thermometers. As for the custom of calling a working thermometer accurate—that’s what it is for.
Eyes and memory can be similarly tested.
Of course, accepting the results of such tests requires acceptance of induction from the past. Maybe the realization you’ve faced “Last Thursdayism” for the first time at undergraduate level is something to cry about, but no one actually does.
Lest I sound too smug, rest assured I am not convinced I would’ve done better before finding Less Wrong.
How do you know that water always boils at the same temperature? Well, you could use a reliable thermometer...
The moral of the story is not so much that science always works, it’s that it works in a way that’s more coherentist than foundationalist. And the downside of coherentism is that you can have more than one equally coherent wordlviews...
Your reliable thermometer doesn’t need to be well-calibrated—it only has to show the same value whenever it’s used to measure boiling water, regardless of what that value is. So the dependence isn’t quite so circular, thankfully.
You are saying it doesn’t need to be accurate. To show that boiling water is always the same temperature, the thermometer doesn’t need to display the rest that temperature …any consistent temperature will do …but it does need to avoid varying randomly, and that is reliability
I see your point. But if water didn’t always boil at the same temperature, why would we bother inventing thermometers?
We have more need to measure the unpredictable than the predictable.
If there was nothing with constant temperature, thermometers would work very differently. My first instinct was to say they wouldn’t work at all. But then I remembered the entire field of economics, so your point stands.
Not every one sees things that way. The more hardline claims require the physical map to exclude others.
Good luck with that. I couldn’t calculate the behaviour of the quarks in a single hydrogen atom if my life depended on it.
Edit: Now I see Sister_Y addressed my point in the very next paragraph, so this entire comment is a reading comprehension fail more than anything.
Necroing:
Go to the beach, light a fire, boil some water. Put the thermometer in the boiled water—does it show 100 degrees Celsius? Still at sea level, put a cup of water in a fridge untill it starts freezing. Put the thermometer in the cup—does it show 0 degrees Celsius?
If yes to both, you have a working thermometer. This way, you don’t rely on the consensus of other thermometers. As for the custom of calling a working thermometer accurate—that’s what it is for.
Eyes and memory can be similarly tested.
Of course, accepting the results of such tests requires acceptance of induction from the past. Maybe the realization you’ve faced “Last Thursdayism” for the first time at undergraduate level is something to cry about, but no one actually does.
Lest I sound too smug, rest assured I am not convinced I would’ve done better before finding Less Wrong.
How do you know that water always boils at the same temperature? Well, you could use a reliable thermometer...
The moral of the story is not so much that science always works, it’s that it works in a way that’s more coherentist than foundationalist. And the downside of coherentism is that you can have more than one equally coherent wordlviews...
Your reliable thermometer doesn’t need to be well-calibrated—it only has to show the same value whenever it’s used to measure boiling water, regardless of what that value is. So the dependence isn’t quite so circular, thankfully.
You are saying it doesn’t need to be accurate. To show that boiling water is always the same temperature, the thermometer doesn’t need to display the rest that temperature …any consistent temperature will do …but it does need to avoid varying randomly, and that is reliability
I remember reading it somewhere...
I see your point. But if water didn’t always boil at the same temperature, why would we bother inventing thermometers?
Right. And since science does work, coherentism gets a big boost in probability, right until the sun stop rising every day.
But would they work equally well? We value science primarily for giving us results, not for being coherent.
If both views are equally coherent and give us equal result (or the results are unclear as of yet), choosing one would be privileging the hipotesis.
We have more need to measure the unpredictable than the predictable.
Not every one sees things that way. The more hardline claims require the physical map to exclude others.
If there was nothing with constant temperature, thermometers would work very differently. My first instinct was to say they wouldn’t work at all. But then I remembered the entire field of economics, so your point stands.
Good luck with that. I couldn’t calculate the behaviour of the quarks in a single hydrogen atom if my life depended on it.