Originally published at: https://boingboing.net/2024/05/29/apparently-seeing-green-apples-is-proof-that-all-ravens-are-black.html
…
4 Likes
6 Likes
Fucking philosophy, eh? How does it work?
6 Likes
Well… if the apple was black it would not be a “green apple.” So that’s good.
3 Likes
Okay, so the headline says “proof” but the body of the article says “evidence”. These are two very different things.
Of course, while finding a non-black non-raven may update your thinking about the truth of the statement “all ravens are black” it only does so in a very tiny, almost unnoticeable way.
Perhaps the headline should really read “Philosopher discovers Bayesian Statistics”.
And of course, since I’ve mentioned that, we must also be careful to use this:
27 Likes
18 Likes
Not if you’re color blind.
8 Likes
Umm - and what about green ravens?
(Did Hempel actually define ‘raven’?)
5 Likes
Yeah, 1 in 30,000 is rare, but hardly unheard of.
8 Likes
All ravens are black?
6 Likes
spoiler: pinocchio gots, at fewest, one hat.
(“spent all me time t’inking of Brazilians who haf left me puzzled”)
3 Likes
This is where my limited grasp of mathematics, and any degree of useful intuition, hit the rocks; so I’m curious whether there are any limitations to how generally this holds: if the set of objects under consideration is not finite; or if every object in the set is black; or we are if there are no ravens in the set.
If I run through a scenario with a relatively well behaved finite pile of objects of various colors and raven/non-raven status I can see the green apple as one step toward a proof by exhaustion; but I kind of washed out of number theory before they had us doing too many proofs on other-than-finite sets; especially the increasingly large and alien ones.
1 Like
On the flipside, if you had an infinite set and you even simplified it to just black ravens, if your largest sample of observation yields only black ravens you still couldn’t definitely say that “all ravens are black” because somewhere in that infinite set there could be an outlier.
What are we talking about again? Why do i smell toast?
4 Likes
IANAPNYAL (philosopher nor yet a logician), so perhaps you can tell me if this applies.
My hypothesis, which is mine and belongs to me, is as follows: there are no red ravens.
To test my hypothesis, I observe ravens — many ravens. Hundreds, thousands of ravens. All the ravens I observe are black. Each black raven I observe is further evidence that my hypothesis is correct.
Then I see a green raven. Logically, this is one more data point confirming my hypothesis. And yet I am shaken to the core, forced to question my entire worldview.
I’ve always thought of this as an example of the problem of inductive vs deductive reasoning, but is it related to Bayesian statistics?
4 Likes
Thank you. Now I can toss my whole hypothesis and go have a drink.
4 Likes
Well how is one establishing what color the ravens are? The feathers are (normally) pigmented black but their feather structure also scatters light making them iridescent.
5 Likes
Now I’m wondering whether painting a black raven red makes it a red raven, or if it’s still a black raven, just painted red.
6 Likes
