[Script Info]
Title: 
[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
Dialogue: 0,0:00:00.04,0:00:05.92,Default,,0000,0000,0000,,(intro music)
Dialogue: 0,0:00:05.92,0:00:07.50,Default,,0000,0000,0000,,My name is Marc Lange.
Dialogue: 0,0:00:07.50,0:00:10.97,Default,,0000,0000,0000,,I teach at the University of\NNorth Carolina at Chapel Hill,
Dialogue: 0,0:00:10.97,0:00:14.47,Default,,0000,0000,0000,,and today I want to talk to you about\Nthe paradox of confirmation.
Dialogue: 0,0:00:14.47,0:00:16.83,Default,,0000,0000,0000,,It's also known as the\N"paradox of the ravens,"
Dialogue: 0,0:00:16.83,0:00:20.60,Default,,0000,0000,0000,,because the philosopher Karl Hempel,\Nwho discovered the paradox,
Dialogue: 0,0:00:20.60,0:00:24.34,Default,,0000,0000,0000,,first presented it in terms of\Nan example involving ravens.
Dialogue: 0,0:00:24.34,0:00:29.95,Default,,0000,0000,0000,,The paradox concerns confirmation, that\Nis, the way that hypotheses in science
Dialogue: 0,0:00:29.95,0:00:33.18,Default,,0000,0000,0000,,and in everyday life are supported\Nby our observations.
Dialogue: 0,0:00:33.18,0:00:38.20,Default,,0000,0000,0000,,As we all know from detective stories,\Na detective gathers evidence for or
Dialogue: 0,0:00:38.20,0:00:42.73,Default,,0000,0000,0000,,against various hypotheses about who\Ncommitted some dastardly crime.
Dialogue: 0,0:00:42.73,0:00:45.38,Default,,0000,0000,0000,,Typically, none of the individual pieces
Dialogue: 0,0:00:45.38,0:00:47.50,Default,,0000,0000,0000,,of evidence available to the detective
Dialogue: 0,0:00:47.50,0:00:51.38,Default,,0000,0000,0000,,is enough all by itself\Nto prove which suspect
Dialogue: 0,0:00:51.38,0:00:53.14,Default,,0000,0000,0000,,did or did not commit the crime.
Dialogue: 0,0:00:53.14,0:00:57.34,Default,,0000,0000,0000,,Instead, a piece of evidence\Nmight count to some degree in
Dialogue: 0,0:00:57.34,0:01:00.69,Default,,0000,0000,0000,,favor of the hypothesis\Nthat the butler is guilty.
Dialogue: 0,0:01:00.69,0:01:04.23,Default,,0000,0000,0000,,The evidence is then said\Nto confirm the hypothesis.
Dialogue: 0,0:01:04.23,0:01:08.24,Default,,0000,0000,0000,,It might confirm the hypothesis\Nstrongly or only to a slight degree.
Dialogue: 0,0:01:08.24,0:01:11.27,Default,,0000,0000,0000,,On the other hand, the\Npiece of evidence might,
Dialogue: 0,0:01:11.27,0:01:14.46,Default,,0000,0000,0000,,to some degree, count against\Nthe truth of the hypothesis.
Dialogue: 0,0:01:14.46,0:01:18.47,Default,,0000,0000,0000,,In that case, the evidence is said\Nto disconfirm the hypothesis.
Dialogue: 0,0:01:18.47,0:01:21.77,Default,,0000,0000,0000,,Again, the disconfirmation\Nmight be strong or weak.
Dialogue: 0,0:01:21.77,0:01:25.37,Default,,0000,0000,0000,,The final possibility is that\Nthe evidence is neutral,
Dialogue: 0,0:01:25.37,0:01:29.36,Default,,0000,0000,0000,,neither confirming nor disconfirming\Nthe hypothesis to any degree.
Dialogue: 0,0:01:29.36,0:01:33.29,Default,,0000,0000,0000,,The paradox of confirmation\Nis concerned with the question
Dialogue: 0,0:01:33.29,0:01:37.71,Default,,0000,0000,0000,,"what does it take for some piece of\Nevidence to confirm a hypothesis,
Dialogue: 0,0:01:37.71,0:01:41.06,Default,,0000,0000,0000,,"rather than to disconfirm it\Nor to be neutral regarding it?"
Dialogue: 0,0:01:41.06,0:01:44.20,Default,,0000,0000,0000,,The paradox of confirmation begins with
Dialogue: 0,0:01:44.20,0:01:48.03,Default,,0000,0000,0000,,three very plausible ideas,\Nand derives from them
Dialogue: 0,0:01:48.03,0:01:51.50,Default,,0000,0000,0000,,a very implausible-looking\Nconclusion about confirmation.
Dialogue: 0,0:01:51.50,0:01:55.65,Default,,0000,0000,0000,,Let's start with the first of these\Nthree plausible-looking ideas,
Dialogue: 0,0:01:55.65,0:01:58.12,Default,,0000,0000,0000,,which I'll call "instance confirmation."
Dialogue: 0,0:01:58.12,0:02:01.00,Default,,0000,0000,0000,,Suppose that we're testing a hypothesis like
Dialogue: 0,0:02:01.00,0:02:04.08,Default,,0000,0000,0000,,"all lightning bolts are\Nelectrical discharges,"
Dialogue: 0,0:02:04.08,0:02:08.36,Default,,0000,0000,0000,,or "all human beings have\Nforty-six chromosomes," or
Dialogue: 0,0:02:08.36,0:02:09.86,Default,,0000,0000,0000,,"all ravens are black."
Dialogue: 0,0:02:09.86,0:02:12.99,Default,,0000,0000,0000,,Each of these hypotheses is general,
Dialogue: 0,0:02:12.99,0:02:16.35,Default,,0000,0000,0000,,in that each takes the\Nform "all Fs are G,"
Dialogue: 0,0:02:16.35,0:02:18.77,Default,,0000,0000,0000,,for some F and some G.
Dialogue: 0,0:02:18.77,0:02:23.68,Default,,0000,0000,0000,,Instance confirmation says that if we're\Ntesting a hypothesis of this form,
Dialogue: 0,0:02:23.68,0:02:26.40,Default,,0000,0000,0000,,and we discover a\Nparticular F to be a G,
Dialogue: 0,0:02:26.40,0:02:29.43,Default,,0000,0000,0000,,then this evidence counts,\Nat least to some degree,
Dialogue: 0,0:02:29.43,0:02:32.04,Default,,0000,0000,0000,,in favor of the hypothesis.
Dialogue: 0,0:02:32.04,0:02:35.27,Default,,0000,0000,0000,,I told you this was going to be\Na plausible-sounding idea.
Dialogue: 0,0:02:35.27,0:02:36.78,Default,,0000,0000,0000,,Isn't it plausible?
Dialogue: 0,0:02:36.78,0:02:40.84,Default,,0000,0000,0000,,The second idea is called\Nthe "equivalence condition."
Dialogue: 0,0:02:40.84,0:02:45.99,Default,,0000,0000,0000,,Suppose we have two hypotheses that say\Nexactly the same thing about the world.
Dialogue: 0,0:02:45.99,0:02:48.55,Default,,0000,0000,0000,,in other words, they are equivalent, in
Dialogue: 0,0:02:48.55,0:02:52.37,Default,,0000,0000,0000,,the sense that they must either\Nboth be true or both be false.
Dialogue: 0,0:02:52.37,0:02:56.28,Default,,0000,0000,0000,,For one of them to be true and the\Nother false would be a contradiction
Dialogue: 0,0:02:56.28,0:02:59.46,Default,,0000,0000,0000,,For instance, suppose that one hypothesis
Dialogue: 0,0:02:59.46,0:03:03.63,Default,,0000,0000,0000,,is that all diamonds are made entirely\Nof carbon, and the other hypothesis
Dialogue: 0,0:03:03.63,0:03:07.37,Default,,0000,0000,0000,,is that carbon is what all diamonds\Nare made entirely out of.
Dialogue: 0,0:03:07.37,0:03:10.17,Default,,0000,0000,0000,,These two hypotheses are equivalent.
Dialogue: 0,0:03:10.17,0:03:12.47,Default,,0000,0000,0000,,What the equivalence condition says
Dialogue: 0,0:03:12.47,0:03:16.20,Default,,0000,0000,0000,,is that if two hypotheses\Nare equivalent, then any
Dialogue: 0,0:03:16.20,0:03:19.42,Default,,0000,0000,0000,,evidence confirming one of\Nthem also confirms the other.
Dialogue: 0,0:03:19.42,0:03:22.99,Default,,0000,0000,0000,,this should strike you\Nas a very plausible idea.
Dialogue: 0,0:03:22.99,0:03:27.91,Default,,0000,0000,0000,,Let's focus on our favorite hypothesis:\Nthat all ravens are black.
Dialogue: 0,0:03:27.92,0:03:32.54,Default,,0000,0000,0000,,The third idea is that this hypothesis\Nis equivalent to another hypothesis.
Dialogue: 0,0:03:32.54,0:03:37.72,Default,,0000,0000,0000,,That other hypothesis is a very clumsy\Nway of saying that all ravens are black.
Dialogue: 0,0:03:37.72,0:03:44.11,Default,,0000,0000,0000,,Here it is: that anything that\Nis non-black is non-raven.
Dialogue: 0,0:03:44.11,0:03:49.58,Default,,0000,0000,0000,,Let me try a different way of explaining\Nthe equivalence of these two hypotheses,
Dialogue: 0,0:03:49.58,0:03:52.40,Default,,0000,0000,0000,,just to make sure that\Nwe're all together on this.
Dialogue: 0,0:03:52.40,0:03:57.51,Default,,0000,0000,0000,,The hypothesis that all Ravens are black\Namounts to a hypothesis ruling out
Dialogue: 0,0:03:57.51,0:04:00.90,Default,,0000,0000,0000,,one possibility: a raven that isn't black.
Dialogue: 0,0:04:00.90,0:04:05.45,Default,,0000,0000,0000,,What about the hypothesis that all\Nnon-black things are non-ravens?
Dialogue: 0,0:04:05.45,0:04:09.92,Default,,0000,0000,0000,,It also amounts to a hypothesis\Nruling out one possibility:
Dialogue: 0,0:04:09.92,0:04:13.72,Default,,0000,0000,0000,,a non-black thing that isn't a non-raven.
Dialogue: 0,0:04:13.72,0:04:16.41,Default,,0000,0000,0000,,In other words, a non-black\Nthing that's a raven.
Dialogue: 0,0:04:16.41,0:04:20.33,Default,,0000,0000,0000,,So both hypotheses are equivalent\Nto the same hypothesis:
Dialogue: 0,0:04:20.33,0:04:22.93,Default,,0000,0000,0000,,that there are no non-black Ravens.
Dialogue: 0,0:04:22.93,0:04:26.30,Default,,0000,0000,0000,,Since the two hypotheses are\Nequivalent to the same hypothesis,
Dialogue: 0,0:04:26.30,0:04:27.88,Default,,0000,0000,0000,,they must be equivalent\Nto each other.
Dialogue: 0,0:04:27.88,0:04:34.20,Default,,0000,0000,0000,,Okay, at last, we are ready for\Nthe paradox of confirmation.
Dialogue: 0,0:04:34.20,0:04:37.70,Default,,0000,0000,0000,,Take the hypothesis that all\Nnon-black things are non-ravens.
Dialogue: 0,0:04:37.70,0:04:39.64,Default,,0000,0000,0000,,That's a general hypothesis.
Dialogue: 0,0:04:39.64,0:04:41.89,Default,,0000,0000,0000,,It takes the form "all Fs are G."
Dialogue: 0,0:04:41.89,0:04:45.37,Default,,0000,0000,0000,,So we can apply the instance\Nconfirmation idea to it.
Dialogue: 0,0:04:45.37,0:04:49.71,Default,,0000,0000,0000,,it would be confirmed by the\Ndiscovery of an F that's a G.
Dialogue: 0,0:04:49.71,0:04:53.08,Default,,0000,0000,0000,,For instance, take the red\Nchair that I'm sitting on.
Dialogue: 0,0:04:53.08,0:04:57.20,Default,,0000,0000,0000,,I am very perceptive, and I've\Nnoticed that it's a non-black thing,
Dialogue: 0,0:04:57.20,0:04:59.43,Default,,0000,0000,0000,,and also that it's not a raven.
Dialogue: 0,0:04:59.43,0:05:03.19,Default,,0000,0000,0000,,So the hypothesis that all\Nnon-black things are non-ravens
Dialogue: 0,0:05:03.19,0:05:06.99,Default,,0000,0000,0000,,is confirmed at, least a bit, by\Nmy observation of my chair.
Dialogue: 0,0:05:06.99,0:05:09.18,Default,,0000,0000,0000,,That's what instance confirmation says.
Dialogue: 0,0:05:09.18,0:05:12.38,Default,,0000,0000,0000,,Now let's apply the equivalence condition.
Dialogue: 0,0:05:12.38,0:05:16.44,Default,,0000,0000,0000,,It tells us that any observation\Nconfirming the hypothesis that all
Dialogue: 0,0:05:16.44,0:05:18.56,Default,,0000,0000,0000,,non-black things are non-ravens
Dialogue: 0,0:05:18.56,0:05:21.82,Default,,0000,0000,0000,,automatically confirms any\Nequivalent hypothesis.
Dialogue: 0,0:05:21.82,0:05:25.03,Default,,0000,0000,0000,,And we've got an equivalent\Nhypothesis in mind:
Dialogue: 0,0:05:25.03,0:05:26.45,Default,,0000,0000,0000,,that all ravens are black.
Dialogue: 0,0:05:26.45,0:05:29.45,Default,,0000,0000,0000,,That was our third plausible idea.
Dialogue: 0,0:05:29.45,0:05:35.00,Default,,0000,0000,0000,,So my observation of my chair confirms\Nthat all non-black things are non-ravens,
Dialogue: 0,0:05:35.00,0:05:39.02,Default,,0000,0000,0000,,and thereby confirms the equivalent\Nhypothesis that all ravens are black.
Dialogue: 0,0:05:39.02,0:05:43.57,Default,,0000,0000,0000,,Now that conclusion about confirmation\Nsounds mighty implausible,
Dialogue: 0,0:05:43.57,0:05:48.31,Default,,0000,0000,0000,,that I could confirm a hypothesis about\Nravens simply by looking around my room
Dialogue: 0,0:05:48.31,0:05:51.68,Default,,0000,0000,0000,,and noticing that my chair, not to\Nmention my desk and my
Dialogue: 0,0:05:51.68,0:05:55.81,Default,,0000,0000,0000,,coffee table, that each of them is\Nnon-black and also not a raven.
Dialogue: 0,0:05:55.81,0:05:59.63,Default,,0000,0000,0000,,I can do ornithology while remaining\Nin the comfort of my room.
Dialogue: 0,0:05:59.63,0:06:02.80,Default,,0000,0000,0000,,So here is the challenge that you face.
Dialogue: 0,0:06:02.80,0:06:07.52,Default,,0000,0000,0000,,either one of those three ideas must be\Nfalse, in a way that explains how we
Dialogue: 0,0:06:07.52,0:06:11.15,Default,,0000,0000,0000,,could have arrived at are false\Nconclusion by using that idea,
Dialogue: 0,0:06:11.15,0:06:15.76,Default,,0000,0000,0000,,or the conclusion must not in fact\Nfollow from those three ideas,
Dialogue: 0,0:06:15.76,0:06:19.62,Default,,0000,0000,0000,,or the conclusion must be true,\Neven though it appears to be false.
Dialogue: 0,0:06:19.62,0:06:21.83,Default,,0000,0000,0000,,Those are your only options.
Dialogue: 0,0:06:21.83,0:06:24.89,Default,,0000,0000,0000,,I leave it to you to think about\Nwhich of them is true.