1
00:00:03,040 --> 00:00:06,760
Speaker 1: Welcome to Stuff to Blow Your Mind production of iHeartRadio.

2
00:00:12,920 --> 00:00:15,400
Speaker 2: Hey, welcome to Stuff to Blow Your Mind. My name

3
00:00:15,440 --> 00:00:16,480
Speaker 2: is Robert Lamb.

4
00:00:16,560 --> 00:00:19,279
Speaker 3: And I'm Joe McCormick. And in today's episode, we're going

5
00:00:19,360 --> 00:00:22,000
Speaker 3: to begin a series in which we look at a

6
00:00:22,040 --> 00:00:27,240
Speaker 3: phenomenon studied in psychology, economics, and decision theory which is

7
00:00:27,280 --> 00:00:34,559
Speaker 3: called ambiguity aversion. Ambiguity aversion is when we prefer risks

8
00:00:34,720 --> 00:00:40,320
Speaker 3: with known probabilities over risks with unknown probabilities, even if

9
00:00:40,360 --> 00:00:43,320
Speaker 3: you have no reason for thinking that the unknown risks

10
00:00:43,400 --> 00:00:46,519
Speaker 3: will be worse than the known risks. And we'll get

11
00:00:46,520 --> 00:00:49,960
Speaker 3: to some concrete examples and foundational research on this in

12
00:00:50,000 --> 00:00:52,839
Speaker 3: a bit, but to an approximation, a good way to

13
00:00:52,880 --> 00:00:57,800
Speaker 3: think about this is that ambiguity aversion is the sentiment

14
00:00:58,000 --> 00:01:01,640
Speaker 3: expressed in the saying, better the devil you know than

15
00:01:01,680 --> 00:01:04,840
Speaker 3: the devil you don't. This is an aphorism I feel

16
00:01:04,840 --> 00:01:08,240
Speaker 3: like people often quote when they're about to justify making

17
00:01:08,240 --> 00:01:13,040
Speaker 3: a really bad or indefensible decision. And the whole point

18
00:01:13,200 --> 00:01:15,960
Speaker 3: of this saying is that the devil you know is

19
00:01:16,040 --> 00:01:19,880
Speaker 3: already bad. That's usually implied, and there's no indication that

20
00:01:19,920 --> 00:01:22,600
Speaker 3: the devil you don't know will be worse. It could

21
00:01:22,680 --> 00:01:26,400
Speaker 3: well be a kinder, gentler devil with a less pointy pitchfork.

22
00:01:26,800 --> 00:01:31,520
Speaker 3: But still we often find ourselves thinking, I'd just rather

23
00:01:31,600 --> 00:01:33,720
Speaker 3: stick with the one where I know how pointy it

24
00:01:33,800 --> 00:01:34,440
Speaker 3: is upfront.

25
00:01:35,080 --> 00:01:38,240
Speaker 2: Yeah, it's often a form of indecision, like refused to

26
00:01:38,240 --> 00:01:41,520
Speaker 2: do anything about the current devil situation, but you know,

27
00:01:41,920 --> 00:01:45,480
Speaker 2: at least I know this devil. Next devil could be worse.

28
00:01:45,520 --> 00:01:47,400
Speaker 2: I don't want a worst devil, that's right.

29
00:01:47,960 --> 00:01:50,280
Speaker 3: This is another one of those aphorisms. We were talking

30
00:01:50,280 --> 00:01:53,920
Speaker 3: about this just recently in the Saint Swithin episode. It's

31
00:01:53,960 --> 00:01:57,000
Speaker 3: one of these aphorisms that I think is only wisdom

32
00:01:57,440 --> 00:02:01,360
Speaker 3: given certain assumptions or conditions that are not really part

33
00:02:01,400 --> 00:02:04,480
Speaker 3: of the saying, Like as stated, it could be purely

34
00:02:04,600 --> 00:02:07,560
Speaker 3: terrible advice, like what if the devil you know is

35
00:02:07,640 --> 00:02:10,200
Speaker 3: really bad? And what if going with the devil you

36
00:02:10,240 --> 00:02:12,760
Speaker 3: don't know gives you a really good shot at improving

37
00:02:12,800 --> 00:02:16,679
Speaker 3: your stay in hell? But also I think it does

38
00:02:16,760 --> 00:02:19,360
Speaker 3: have a kind of wisdom if you don't think of

39
00:02:19,400 --> 00:02:23,280
Speaker 3: it as advice, but rather as an ironic observation about

40
00:02:23,400 --> 00:02:25,679
Speaker 3: human nature, in which case I think it's sort of

41
00:02:25,760 --> 00:02:28,680
Speaker 3: dead on this This is often how we think, and

42
00:02:28,720 --> 00:02:33,080
Speaker 3: that is the core observation also of the ambiguity aversion literature.

43
00:02:33,360 --> 00:02:36,160
Speaker 2: Yeah yeah, I mean because you can also ask, well,

44
00:02:36,320 --> 00:02:38,040
Speaker 2: what if we just didn't have a devil? Is that

45
00:02:38,120 --> 00:02:39,919
Speaker 2: on the table? Can we just not have a devil?

46
00:02:39,960 --> 00:02:42,480
Speaker 2: Why do we have to worry about the devil we

47
00:02:42,560 --> 00:02:45,760
Speaker 2: have versus the devil we don't have yet it's you know,

48
00:02:45,800 --> 00:02:48,840
Speaker 2: it can be, and I guess it's rather condemning human

49
00:02:48,919 --> 00:02:51,760
Speaker 2: nature in that aspect as well, where the answer is, well,

50
00:02:51,760 --> 00:02:54,079
Speaker 2: you got to have a devil where humans we make

51
00:02:54,120 --> 00:02:56,280
Speaker 2: our own and you've got to have one in play.

52
00:02:56,400 --> 00:02:58,359
Speaker 3: You should have thought of that before you double park Rob.

53
00:02:58,560 --> 00:02:59,800
Speaker 2: Yeah, now you're here in hell with.

54
00:02:59,800 --> 00:03:00,520
Speaker 3: The rest of us.

55
00:03:00,919 --> 00:03:03,840
Speaker 2: There's a rather keen Chinese saying that goes along these

56
00:03:03,919 --> 00:03:06,840
Speaker 2: lines as well. It's a well known idiom that states

57
00:03:07,200 --> 00:03:10,000
Speaker 2: that it's quote easy to dodge the spear in the open,

58
00:03:10,360 --> 00:03:12,320
Speaker 2: hard to avoid a stab in the dark.

59
00:03:12,639 --> 00:03:16,960
Speaker 3: Oh that's interesting, because yeah, that's the same idea. It

60
00:03:17,120 --> 00:03:21,280
Speaker 3: isn't necessarily true in all circumstances, like in the dark,

61
00:03:21,360 --> 00:03:24,360
Speaker 3: the person stabbing at you also can't see you very well,

62
00:03:24,760 --> 00:03:26,640
Speaker 3: so it might be hard to dodge, but it's also

63
00:03:26,680 --> 00:03:28,880
Speaker 3: hard for them to hit you. But in the dark,

64
00:03:29,280 --> 00:03:31,960
Speaker 3: you just don't really know, you don't really know what's

65
00:03:32,000 --> 00:03:34,000
Speaker 3: going on, so it just seems scarier.

66
00:03:34,600 --> 00:03:37,600
Speaker 2: Yeah. Absolutely, And it reminds me a bit of another

67
00:03:37,720 --> 00:03:40,480
Speaker 2: Chinese saying, better to be a dog in times of

68
00:03:40,480 --> 00:03:44,200
Speaker 2: tranquility than a human in times of chaos, And this

69
00:03:44,280 --> 00:03:47,200
Speaker 2: has some of the same sentiments as a fourteenth century

70
00:03:47,240 --> 00:03:51,920
Speaker 2: Islamic saying that is sometimes misattributed as being of Chinese

71
00:03:51,960 --> 00:03:55,040
Speaker 2: origin itself, and that is better a century of tyranny

72
00:03:55,280 --> 00:03:56,560
Speaker 2: than one day of chaos.

73
00:03:57,560 --> 00:03:59,520
Speaker 3: Also, don't know if I actually agree with that.

74
00:04:00,120 --> 00:04:04,040
Speaker 2: Well, I don't know. It's I think about it a

75
00:04:04,040 --> 00:04:09,920
Speaker 2: lot for various reasons, and I do. I do agree

76
00:04:10,000 --> 00:04:14,520
Speaker 2: with the basic sentiment of it. When you find yourself

77
00:04:14,560 --> 00:04:18,600
Speaker 2: dipping into chaos. And I'm not talking about like capital

78
00:04:18,640 --> 00:04:21,520
Speaker 2: C chaos, like absolute chaos, but these little moments of

79
00:04:21,600 --> 00:04:24,560
Speaker 2: chaos and disorder, you can easily line yourself up with

80
00:04:24,640 --> 00:04:26,719
Speaker 2: the idea of like, well, yeah, this is why people,

81
00:04:27,600 --> 00:04:30,800
Speaker 2: you know, fall for tyranny. This is because they have

82
00:04:30,880 --> 00:04:34,240
Speaker 2: these moments of chaos and uncertainty and they're like, well,

83
00:04:34,279 --> 00:04:37,000
Speaker 2: I would just rather have this sort of known top

84
00:04:37,040 --> 00:04:40,520
Speaker 2: down oppression in place rather than some sort of ambiguous

85
00:04:40,600 --> 00:04:42,000
Speaker 2: chaos bubbling up around me.

86
00:04:42,320 --> 00:04:45,520
Speaker 3: Better are something that's definitely bad than something I don't

87
00:04:45,600 --> 00:04:47,000
Speaker 3: understand and can't predict.

88
00:04:47,480 --> 00:04:51,440
Speaker 2: Yeah, Now, ambiguity and chaos, we should note here, are

89
00:04:51,520 --> 00:04:55,240
Speaker 2: not the same thing. Ambiguity is a lack of clarity. Well,

90
00:04:55,360 --> 00:04:58,400
Speaker 2: chaos is a state of disorder, and the former may

91
00:04:58,440 --> 00:05:01,520
Speaker 2: well lead into the other. And chaos may be seen

92
00:05:01,560 --> 00:05:05,080
Speaker 2: as a state of absolute ambiguity and certainty as well.

93
00:05:05,120 --> 00:05:07,800
Speaker 2: So in that respect, it becomes even more difficult when

94
00:05:07,800 --> 00:05:11,600
Speaker 2: we try and weigh chaos and ambiguity in our concerns

95
00:05:11,640 --> 00:05:12,400
Speaker 2: about the future.

96
00:05:12,640 --> 00:05:15,719
Speaker 3: I'm just trying to think of familiar, everyday examples of

97
00:05:15,800 --> 00:05:21,000
Speaker 3: ambiguity aversion, and I think I came across one internally

98
00:05:21,200 --> 00:05:26,160
Speaker 3: in the now commonly acknowledged sentiment of not wanting to

99
00:05:26,200 --> 00:05:30,400
Speaker 3: go out and try anything you know, So, like, do

100
00:05:30,400 --> 00:05:33,880
Speaker 3: you ever find yourself trying to find a way to

101
00:05:34,080 --> 00:05:38,320
Speaker 3: avoid going through an experience that is new and unfamiliar

102
00:05:38,680 --> 00:05:41,640
Speaker 3: and thus unpredictable to you? So it seems like, I

103
00:05:41,680 --> 00:05:44,040
Speaker 3: don't know, it just seems kind of fraught, like a hassle,

104
00:05:44,560 --> 00:05:47,520
Speaker 3: And instead of doing that, choosing to have an experience

105
00:05:47,600 --> 00:05:50,720
Speaker 3: that you pretty much know that you will probably not enjoy.

106
00:05:54,160 --> 00:05:56,320
Speaker 2: Yeah, I think I know what you're saying here. Yeah,

107
00:05:56,360 --> 00:06:00,960
Speaker 2: I mean, I mean life is perpetually a situation where

108
00:06:01,200 --> 00:06:03,599
Speaker 2: it's like, am I going to stick with what's comfortable

109
00:06:03,720 --> 00:06:05,919
Speaker 2: or am I going to say yes to adventure? Am

110
00:06:05,960 --> 00:06:11,480
Speaker 2: I going to try something new? And I often find myself,

111
00:06:11,720 --> 00:06:14,440
Speaker 2: you know, at the crossroads of that moment, because there's

112
00:06:14,480 --> 00:06:16,479
Speaker 2: a huge part of me that doesn't want to try

113
00:06:16,520 --> 00:06:21,360
Speaker 2: new things, and I get anxious about the various experiences

114
00:06:21,400 --> 00:06:25,200
Speaker 2: aligned with those kinds of quests. But on the other hand,

115
00:06:25,279 --> 00:06:27,599
Speaker 2: I have to recognize that like most of the really

116
00:06:28,400 --> 00:06:31,279
Speaker 2: fulfilling things in my life have been because I said

117
00:06:31,320 --> 00:06:34,000
Speaker 2: yes to adventure and because I took a chance on

118
00:06:34,040 --> 00:06:37,839
Speaker 2: something and I went out of my comfort zone. But

119
00:06:37,880 --> 00:06:40,200
Speaker 2: then again, we call it our comfort zone because it

120
00:06:40,240 --> 00:06:42,400
Speaker 2: is comfortable. It is a nice place, it is a

121
00:06:42,440 --> 00:06:45,360
Speaker 2: comforting place. So it's you know, it's given take.

122
00:06:45,640 --> 00:06:48,359
Speaker 3: Yes, I think that's absolutely correct. But I think the

123
00:06:48,400 --> 00:06:51,200
Speaker 3: really weird thing is that sometimes your quote comfort zone

124
00:06:51,360 --> 00:06:55,280
Speaker 3: isn't comfortable. I mean, sometimes we prefer an experience that's

125
00:06:55,360 --> 00:06:57,920
Speaker 3: not even very nice. It's still like something that we

126
00:06:57,960 --> 00:07:00,240
Speaker 3: know we won't like, but it's just it's you know,

127
00:07:00,279 --> 00:07:02,640
Speaker 3: it's familiar. We at least know what we're getting with it,

128
00:07:03,000 --> 00:07:05,640
Speaker 3: over something that is just question marks.

129
00:07:05,880 --> 00:07:10,760
Speaker 2: Right right, will choose a known okayness over a potentially

130
00:07:11,000 --> 00:07:12,160
Speaker 2: amazing experience.

131
00:07:12,640 --> 00:07:16,880
Speaker 3: So one of the most important early writings on ambiguity

132
00:07:16,920 --> 00:07:20,880
Speaker 3: aversion is a classic paper in the Quarterly Journal of

133
00:07:20,920 --> 00:07:25,920
Speaker 3: Economics from nineteen sixty one by the American economist, whistleblower

134
00:07:26,000 --> 00:07:30,360
Speaker 3: and political activist Daniel Ellsberg. The paper was called Risk,

135
00:07:30,560 --> 00:07:35,240
Speaker 3: Ambiguity and the Savage Axioms. That's savage with a capital S.

136
00:07:35,400 --> 00:07:38,720
Speaker 3: That's a person's name. It's not describing the axioms as savage.

137
00:07:39,920 --> 00:07:42,680
Speaker 3: And just to go ahead and mention it now because

138
00:07:42,720 --> 00:07:45,360
Speaker 3: I also use this as a source in writing about

139
00:07:46,000 --> 00:07:50,320
Speaker 3: Elsberg and this paper's impact. It was also reading a

140
00:07:50,480 --> 00:07:54,920
Speaker 3: paper called Ambiguity and Ambiguity Aversion in the Handbook of

141
00:07:54,960 --> 00:07:58,080
Speaker 3: the Economics of Risk and Uncertainty. This is by Mark J.

142
00:07:58,320 --> 00:08:03,960
Speaker 3: Maschina and Marciano Cinneskalchi, published by north Holland in twenty fourteen.

143
00:08:04,280 --> 00:08:07,120
Speaker 3: This is a sort of general overview of the concept

144
00:08:07,160 --> 00:08:09,960
Speaker 3: of ambiguity aversion and a survey of a bunch of

145
00:08:10,000 --> 00:08:13,000
Speaker 3: different models of it. Now, before we get into the

146
00:08:13,040 --> 00:08:16,400
Speaker 3: idea that Elsburg explored in this very important paper, robbed it.

147
00:08:16,480 --> 00:08:18,280
Speaker 3: You had a bit of bio in Elsburg, didn't you.

148
00:08:18,960 --> 00:08:23,600
Speaker 2: Yeah. Yeah. Daniel Ellsberg quite an interesting fellow of nineteen

149
00:08:23,640 --> 00:08:28,960
Speaker 2: thirty one through twenty twenty three. In addition to Elsberg's paradox,

150
00:08:29,040 --> 00:08:33,120
Speaker 2: which we're getting into here, is he's also famous, perhaps

151
00:08:33,200 --> 00:08:36,440
Speaker 2: more so in a general sense, as the Rand Corporation

152
00:08:36,559 --> 00:08:39,520
Speaker 2: employee who photocopied and released what would come to be

153
00:08:39,600 --> 00:08:43,240
Speaker 2: known as the Pentagon Papers. This is a United States

154
00:08:43,280 --> 00:08:46,960
Speaker 2: Department of Defense history of the US's political and military

155
00:08:47,000 --> 00:08:50,520
Speaker 2: involvement in Vietnam from nineteen forty five through nineteen sixty eight.

156
00:08:51,040 --> 00:08:53,360
Speaker 3: The key point being that Ellsberg believed that getting these

157
00:08:53,360 --> 00:08:57,200
Speaker 3: papers out there was important in informing the public about

158
00:08:57,559 --> 00:08:59,840
Speaker 3: things that they didn't know about how the Vietnam War

159
00:09:00,160 --> 00:09:02,640
Speaker 3: being prosecuted and represented right right.

160
00:09:02,720 --> 00:09:06,840
Speaker 2: It was very much a whistle blowing act on Ellsberg's part,

161
00:09:07,320 --> 00:09:10,000
Speaker 2: and these papers were published by the New York Times

162
00:09:10,000 --> 00:09:12,680
Speaker 2: in nineteen seventy one. This would be a decade after

163
00:09:13,080 --> 00:09:17,280
Speaker 2: the publication regarding the Elsberg paradox. And yeah, they detailed

164
00:09:17,400 --> 00:09:20,679
Speaker 2: secret developments about the scope of the US military efforts

165
00:09:20,679 --> 00:09:25,280
Speaker 2: in Vietnam, the real sticking points being the deception involving

166
00:09:25,320 --> 00:09:28,680
Speaker 2: the Gulf of Tonkin incident, unreported expansion of the war,

167
00:09:28,800 --> 00:09:31,640
Speaker 2: the true objective of the war post nineteen sixty five,

168
00:09:31,800 --> 00:09:36,040
Speaker 2: which basically was more about containing China and preventing a

169
00:09:36,120 --> 00:09:41,640
Speaker 2: humiliating US defeat rather than any specific objectives to help

170
00:09:41,679 --> 00:09:44,280
Speaker 2: the people of Vietnam, and the lack of a clear

171
00:09:44,320 --> 00:09:47,320
Speaker 2: plan for victory as well. So I mentioned a lot

172
00:09:47,360 --> 00:09:49,400
Speaker 2: of you have at least heard of the Pentagon papers,

173
00:09:49,400 --> 00:09:52,000
Speaker 2: even if you're not familiar with the full story, because

174
00:09:52,480 --> 00:09:56,760
Speaker 2: this incident this led to conspiracy and espionage charges against

175
00:09:56,760 --> 00:10:02,400
Speaker 2: Ellsberg that were then later dismissed, greatly fueled the anti

176
00:10:02,440 --> 00:10:04,920
Speaker 2: war movement as well as distrust in the government at

177
00:10:04,920 --> 00:10:08,120
Speaker 2: the time, and led to the Nixon Whitehouse's use of

178
00:10:08,120 --> 00:10:11,600
Speaker 2: the infamous White House Plumbers to discredit Elsberg. And of course,

179
00:10:11,679 --> 00:10:14,000
Speaker 2: all of this ends up feeding into the Watergate scandal,

180
00:10:14,040 --> 00:10:17,559
Speaker 2: which is another one of those major historical touchstones from

181
00:10:17,720 --> 00:10:20,880
Speaker 2: the time period that again, if you're not super familiar

182
00:10:20,880 --> 00:10:24,000
Speaker 2: with Watergate, you've at least heard of Watergate, and you've

183
00:10:24,000 --> 00:10:28,640
Speaker 2: heard about the other different gates out there that allude

184
00:10:28,679 --> 00:10:30,360
Speaker 2: back to what a scandal it was.

185
00:10:31,440 --> 00:10:33,680
Speaker 3: Now everything gets a gate, suffix.

186
00:10:33,520 --> 00:10:37,880
Speaker 2: Right, no matter how deserve, any or undeserved now. Elsberg

187
00:10:37,960 --> 00:10:42,040
Speaker 2: remained an outspoken activist and whistleblower throughout his life, particularly

188
00:10:42,720 --> 00:10:47,240
Speaker 2: regarding the US's military actions under multiple presidents, including the

189
00:10:47,679 --> 00:10:50,960
Speaker 2: current one, the Russian invasion of Ukraine, and also in

190
00:10:51,000 --> 00:10:54,240
Speaker 2: support of various other whistle blowers. He was a vocal

191
00:10:54,240 --> 00:10:58,160
Speaker 2: opponent of the use and the threat of nuclear weapons

192
00:10:58,520 --> 00:11:01,960
Speaker 2: and the rhetoric around their use, and he is sometimes

193
00:11:02,000 --> 00:11:06,040
Speaker 2: described as a nuclear war planner turned nuclear war activists,

194
00:11:06,480 --> 00:11:11,080
Speaker 2: and his work at the Rand Corporation did involve such contemplations. Yeah.

195
00:11:11,120 --> 00:11:13,400
Speaker 3: I think a big part of his pitch in his

196
00:11:13,760 --> 00:11:17,280
Speaker 3: anti war and anti nuclear weapons activism was like, look,

197
00:11:17,320 --> 00:11:21,120
Speaker 3: I've been there where the power players are executing you know,

198
00:11:21,200 --> 00:11:24,719
Speaker 3: global military and nuclear strategy. You should be worried. This

199
00:11:24,840 --> 00:11:26,320
Speaker 3: is things are not okay.

200
00:11:27,320 --> 00:11:31,199
Speaker 2: Yeah. One of his major overarching criticisms of nuclear weapons,

201
00:11:31,280 --> 00:11:33,439
Speaker 2: and I feel like we're probably preaching to the choir

202
00:11:33,600 --> 00:11:35,240
Speaker 2: on this one. I can't imagine we have any big

203
00:11:35,480 --> 00:11:39,199
Speaker 2: nuclear war fans out there, was that their use is

204
00:11:39,240 --> 00:11:42,720
Speaker 2: a deterrent one of the major positive spins that are

205
00:11:42,760 --> 00:11:47,360
Speaker 2: often given about having a vast nuclear arsenal. This idea

206
00:11:47,440 --> 00:11:51,120
Speaker 2: of the nuclear deterrent was itself highly unreliable and unstable.

207
00:11:51,440 --> 00:11:55,960
Speaker 2: There were simply too many variables human, technological, situational that

208
00:11:56,040 --> 00:12:01,880
Speaker 2: could lead to not only catastrophic but existential unintended consequences.

209
00:12:02,280 --> 00:12:04,280
Speaker 3: Yeah, too many ways for it to go wrong, and

210
00:12:04,360 --> 00:12:06,880
Speaker 3: if it does go wrong, the consequences are too great.

211
00:12:07,320 --> 00:12:09,800
Speaker 2: By the way Elsberg has been portrayed, I believe twice

212
00:12:09,960 --> 00:12:12,800
Speaker 2: in major motion pictures, once by James Spader and once

213
00:12:12,840 --> 00:12:22,240
Speaker 2: by Matthew Reese.

214
00:12:24,080 --> 00:12:27,960
Speaker 3: So coming back to the Elsberg paper from nineteen sixty one, again,

215
00:12:27,960 --> 00:12:30,600
Speaker 3: that's risk ambiguity in the Savage axioms. This is the

216
00:12:30,640 --> 00:12:32,960
Speaker 3: really important one. I'd look it up. I think it

217
00:12:33,040 --> 00:12:36,959
Speaker 3: had like, I don't know, like tens of thousands of citations.

218
00:12:37,000 --> 00:12:41,280
Speaker 3: It's like a very, very famous and influential economics paper.

219
00:12:41,920 --> 00:12:45,240
Speaker 3: In this paper he illustrates the idea of ambiguity a

220
00:12:45,320 --> 00:12:49,000
Speaker 3: version with a number of thought experiments, and this is

221
00:12:49,040 --> 00:12:52,320
Speaker 3: one of them. So this is the three color urn.

222
00:12:53,559 --> 00:12:56,840
Speaker 3: Imagine that there's an urn like a jar that is

223
00:12:56,920 --> 00:13:00,600
Speaker 3: filled with ninety little ping pong balls that are a

224
00:13:00,760 --> 00:13:04,400
Speaker 3: mix of different colors. You cannot see inside the urn,

225
00:13:04,440 --> 00:13:06,840
Speaker 3: it's opaque, But you're going to play a game where

226
00:13:06,880 --> 00:13:09,360
Speaker 3: you reach into the urn and you pull out one

227
00:13:09,440 --> 00:13:12,720
Speaker 3: ball at a time, and you can win prizes based

228
00:13:12,760 --> 00:13:14,880
Speaker 3: on the color of the ball that you pull out.

229
00:13:15,559 --> 00:13:18,880
Speaker 3: You know that thirty of the balls inside the urn

230
00:13:19,080 --> 00:13:22,000
Speaker 3: are red, So since thirty out of ninety are red,

231
00:13:22,040 --> 00:13:24,760
Speaker 3: there's a one in three chance that a random ball

232
00:13:24,840 --> 00:13:27,000
Speaker 3: is going to be red. The odds there are pretty clear.

233
00:13:27,360 --> 00:13:31,120
Speaker 3: The other sixty balls, however, are a mix of yellow

234
00:13:31,240 --> 00:13:34,640
Speaker 3: and black. And here's the really important part. You don't

235
00:13:34,720 --> 00:13:39,000
Speaker 3: know the composition of that group. So those sixty balls

236
00:13:39,040 --> 00:13:41,839
Speaker 3: could be thirty yellow and thirty black. It could be

237
00:13:41,960 --> 00:13:45,199
Speaker 3: sixty yellow and zero black. It could be sixty black

238
00:13:45,200 --> 00:13:49,160
Speaker 3: and zero yellow, or any mix in between. Since you

239
00:13:49,240 --> 00:13:52,520
Speaker 3: don't know the mix, you don't know the probability of

240
00:13:52,600 --> 00:13:55,559
Speaker 3: drawing a yellow ball or a black ball ahead of time.

241
00:13:55,800 --> 00:13:59,040
Speaker 3: For red, it's one in three. For either yellow or black,

242
00:13:59,240 --> 00:14:03,719
Speaker 3: it's anywhere between zero and two out of three. Now,

243
00:14:03,760 --> 00:14:06,320
Speaker 3: imagine it is your turn to draw a ball out

244
00:14:06,320 --> 00:14:09,280
Speaker 3: of the urn, and you can take one of two bets.

245
00:14:09,840 --> 00:14:13,400
Speaker 3: Bet a you get one hundred dollars if your ball

246
00:14:13,520 --> 00:14:16,080
Speaker 3: is red, So if you pick this option, your chance

247
00:14:16,120 --> 00:14:19,560
Speaker 3: of winning one hundred dollars is one in three. In

248
00:14:19,640 --> 00:14:22,320
Speaker 3: bet B you get one hundred dollars if your ball

249
00:14:22,440 --> 00:14:25,960
Speaker 3: is black. Since you don't know the mix of yellow

250
00:14:26,040 --> 00:14:29,000
Speaker 3: and black, you don't know your odds of winning, but

251
00:14:29,080 --> 00:14:32,680
Speaker 3: they are somewhere between zero and two out of three,

252
00:14:33,080 --> 00:14:35,400
Speaker 3: with the midpoint being one out of three, which is

253
00:14:35,440 --> 00:14:38,240
Speaker 3: the same as the odds on red. Which one do

254
00:14:38,280 --> 00:14:38,680
Speaker 3: you pick?

255
00:14:39,360 --> 00:14:42,960
Speaker 2: Well, it sounds like bet A is the suer thing, right.

256
00:14:43,360 --> 00:14:45,760
Speaker 3: If somebody offered me this game, I would probably pick

257
00:14:45,840 --> 00:14:48,360
Speaker 3: bet A. Yeah, it just seems safer. It's like, okay,

258
00:14:48,560 --> 00:14:50,520
Speaker 3: I know what I'm dealing with. There one in three

259
00:14:51,200 --> 00:14:53,680
Speaker 3: the other option. Your odds could be better than one

260
00:14:53,680 --> 00:14:55,400
Speaker 3: in three, but you you just don't know.

261
00:14:56,000 --> 00:14:56,720
Speaker 2: They could be worse.

262
00:14:56,760 --> 00:14:59,600
Speaker 3: It could be zero, They could be worse, could be better.

263
00:14:59,680 --> 00:15:03,480
Speaker 3: You have no way of knowing either way, Ellsberg suggested,

264
00:15:03,640 --> 00:15:08,600
Speaker 3: and later experiments would empirically demonstrate that most people would

265
00:15:08,600 --> 00:15:11,760
Speaker 3: bet on red instead of black, even though there is

266
00:15:11,840 --> 00:15:15,680
Speaker 3: no objective reason to think that red is a better bet.

267
00:15:16,080 --> 00:15:19,880
Speaker 3: Following the pick red strategy will not necessarily earn you

268
00:15:19,960 --> 00:15:22,960
Speaker 3: more money in this game, But people really do, in fact,

269
00:15:23,240 --> 00:15:26,840
Speaker 3: just like it better to pick red. It feels right,

270
00:15:26,960 --> 00:15:29,600
Speaker 3: that's just what we think we should do. That in

271
00:15:29,680 --> 00:15:32,600
Speaker 3: itself is kind of weird and interesting, like why do

272
00:15:32,640 --> 00:15:36,440
Speaker 3: we prefer to avoid ambiguity even when there is no

273
00:15:36,680 --> 00:15:40,800
Speaker 3: clear advantage in doing so. Then there's a really interesting wrinkle.

274
00:15:41,120 --> 00:15:43,800
Speaker 3: Imagine you're playing the same game. You are drawing a

275
00:15:43,840 --> 00:15:47,960
Speaker 3: ball out of an urn, and this time you're given

276
00:15:47,960 --> 00:15:50,760
Speaker 3: assurance that the mix of the balls is the same

277
00:15:50,800 --> 00:15:53,040
Speaker 3: as it was the first time. So whatever was the

278
00:15:53,080 --> 00:15:55,640
Speaker 3: case with the mix of yellow and black in there,

279
00:15:56,200 --> 00:15:57,720
Speaker 3: it's going to be the same as it was with

280
00:15:57,760 --> 00:15:59,800
Speaker 3: your last draw. And you get a chance to make

281
00:15:59,800 --> 00:16:03,560
Speaker 3: a second bet. You can take either bet C, where

282
00:16:03,600 --> 00:16:06,080
Speaker 3: you get one hundred dollars if the ball drawn out

283
00:16:06,200 --> 00:16:11,240
Speaker 3: is either yellow or red, and this one has an

284
00:16:11,400 --> 00:16:14,080
Speaker 3: unknown chance of winning, because of course you don't know

285
00:16:14,120 --> 00:16:16,680
Speaker 3: how many yellow or black balls there are. You know

286
00:16:16,800 --> 00:16:19,480
Speaker 3: that your chance of winning is somewhere between one and

287
00:16:19,560 --> 00:16:22,240
Speaker 3: three if there are you know, the thirty red balls

288
00:16:22,240 --> 00:16:25,160
Speaker 3: and zero yellow balls, or one hundred percent if there

289
00:16:25,160 --> 00:16:28,000
Speaker 3: are sixty yellow balls, Or you can take bet D

290
00:16:28,600 --> 00:16:30,920
Speaker 3: where you get one hundred dollars. If the ball is

291
00:16:31,000 --> 00:16:35,160
Speaker 3: either yellow or black, this one has a guaranteed two

292
00:16:35,280 --> 00:16:38,480
Speaker 3: thirds chance to win, because we know that together yellow

293
00:16:38,560 --> 00:16:40,800
Speaker 3: and black make sixty of the ninety balls that we

294
00:16:40,840 --> 00:16:44,160
Speaker 3: don't know what the mix is. In this second trial,

295
00:16:44,640 --> 00:16:48,560
Speaker 3: people will tend to pick bet D once again, because

296
00:16:48,600 --> 00:16:51,680
Speaker 3: the probabilities are known. With D, it's a guaranteed two

297
00:16:51,680 --> 00:16:55,240
Speaker 3: thirds chance to win. With bets C, it's somewhere between

298
00:16:55,360 --> 00:16:59,080
Speaker 3: one one in three and one hundred percent. Picking bet

299
00:16:59,160 --> 00:17:03,000
Speaker 3: D allows them to avoid placing bets C, which, again,

300
00:17:03,080 --> 00:17:07,399
Speaker 3: which has ambiguous chances. Here's the really interesting thing that

301
00:17:08,119 --> 00:17:13,560
Speaker 3: Elsberg pointed out. There's actually a contradiction in people's behavior

302
00:17:13,960 --> 00:17:17,680
Speaker 3: between the first and second round of this game. Assuming

303
00:17:17,720 --> 00:17:19,879
Speaker 3: you are actually doing your best to win the money.

304
00:17:20,359 --> 00:17:23,679
Speaker 3: In round one, if you bet on red instead of black,

305
00:17:24,119 --> 00:17:27,639
Speaker 3: that implies you believe that less than thirty of the

306
00:17:27,760 --> 00:17:31,239
Speaker 3: unknown sixty balls are black. Right, That makes you it

307
00:17:31,280 --> 00:17:33,840
Speaker 3: implies that red has a better chance of winning, so

308
00:17:34,400 --> 00:17:36,960
Speaker 3: less than half of them have to be black. However,

309
00:17:37,000 --> 00:17:39,560
Speaker 3: in the second round, if you take the second option

310
00:17:39,920 --> 00:17:42,879
Speaker 3: picking yellow and black, it implies that you think that

311
00:17:43,359 --> 00:17:46,840
Speaker 3: more than thirty of the unknown balls are black. Otherwise

312
00:17:46,880 --> 00:17:50,119
Speaker 3: the red and yellow combination option would have the better

313
00:17:50,240 --> 00:17:56,120
Speaker 3: chance of winning. Right, These two bets in combination make

314
00:17:56,280 --> 00:18:02,000
Speaker 3: no sense because they imply self contradictory beliefs. It says

315
00:18:02,040 --> 00:18:05,200
Speaker 3: that I believe that more balls are black and less

316
00:18:05,200 --> 00:18:08,919
Speaker 3: balls are black at the same time. Obviously that can't

317
00:18:08,960 --> 00:18:12,760
Speaker 3: really be the case. But there could be another issue

318
00:18:12,880 --> 00:18:18,280
Speaker 3: driving this behavior, which is simply that people don't like ambiguity,

319
00:18:18,600 --> 00:18:22,000
Speaker 3: and they will pay up for known odds, even if

320
00:18:22,040 --> 00:18:27,640
Speaker 3: doing so implies mutually exclusive assumptions. I was trying to think,

321
00:18:27,800 --> 00:18:29,400
Speaker 3: you know, as I said a minute ago, I think

322
00:18:29,440 --> 00:18:32,280
Speaker 3: I would probably pick red if given the option to

323
00:18:32,320 --> 00:18:35,920
Speaker 3: play this game. Obviously, I sort of already knew the

324
00:18:35,960 --> 00:18:37,879
Speaker 3: deal here, so I can't know what I would do

325
00:18:38,080 --> 00:18:41,120
Speaker 3: going into it blind like most people, but I think

326
00:18:41,160 --> 00:18:46,480
Speaker 3: I would, for no rational reason, prefer the uncertainty reducing options.

327
00:18:47,359 --> 00:18:50,480
Speaker 3: I think the most likely reason for that is even

328
00:18:50,520 --> 00:18:52,960
Speaker 3: if I were given assurances that this game was on

329
00:18:53,000 --> 00:18:56,320
Speaker 3: the up and up, I would suspect some kind of trick,

330
00:18:56,520 --> 00:18:59,760
Speaker 3: Like if I'm being asked to bet on uncertainty, There's

331
00:18:59,800 --> 00:19:02,719
Speaker 3: some part of me that just kind of rises up

332
00:19:02,760 --> 00:19:05,239
Speaker 3: and says, uh, there's a scam here that you are

333
00:19:05,280 --> 00:19:05,840
Speaker 3: not seeing.

334
00:19:06,040 --> 00:19:08,280
Speaker 2: Yeah, yeah, I mean you kind of expect that within

335
00:19:08,359 --> 00:19:12,880
Speaker 2: the constraints of a testing environment, and you definitely expect

336
00:19:12,880 --> 00:19:16,399
Speaker 2: that out of the world. You know, whatever form this

337
00:19:16,600 --> 00:19:19,200
Speaker 2: was taking, you'd assume that somebody had a vested interest

338
00:19:19,800 --> 00:19:22,520
Speaker 2: in manipulating how you were going to behave to the

339
00:19:22,960 --> 00:19:24,920
Speaker 2: how you're going to respond to the two possibilities.

340
00:19:25,160 --> 00:19:27,760
Speaker 3: Yeah. Elsberg also in this paper, by the way, he

341
00:19:27,760 --> 00:19:30,000
Speaker 3: came up with a second illustration of the same principle.

342
00:19:30,040 --> 00:19:32,680
Speaker 3: So if you go reading about this, you'll read about

343
00:19:32,680 --> 00:19:35,400
Speaker 3: the three color urn and then the two urn experiment.

344
00:19:35,440 --> 00:19:38,720
Speaker 3: The second illustration used two eurns instead of one, because

345
00:19:38,720 --> 00:19:42,000
Speaker 3: it made the contradiction between the different bets even clearer.

346
00:19:42,840 --> 00:19:47,760
Speaker 3: And this contradictory self contradictory behavior is now called the

347
00:19:47,880 --> 00:19:51,679
Speaker 3: Elsberg paradox. Elsberg himself didn't call it that, that's what

348
00:19:51,840 --> 00:19:56,119
Speaker 3: other people they put his name on it. Anyway, Elsberg's

349
00:19:56,119 --> 00:20:01,520
Speaker 3: paper about ambiguity aversion was not just about observing a

350
00:20:01,560 --> 00:20:05,760
Speaker 3: weird little quirk of human behavior. It was especially of

351
00:20:05,800 --> 00:20:10,600
Speaker 3: interest to the fields of economics and decision theory modeling

352
00:20:10,640 --> 00:20:14,919
Speaker 3: how people make decisions, because it sought to undermine some

353
00:20:15,000 --> 00:20:20,720
Speaker 3: of the major underpinnings of the fields, specifically providing evidence

354
00:20:20,800 --> 00:20:24,800
Speaker 3: against some of the tenets of a framework called subjective

355
00:20:24,960 --> 00:20:29,359
Speaker 3: expected utility theory or savage as axioms, named after the

356
00:20:29,400 --> 00:20:35,640
Speaker 3: American mathematician Leonard Savage who formulated them. Subjective expected utility

357
00:20:35,680 --> 00:20:39,520
Speaker 3: theory or SEU theory is a way of modeling how

358
00:20:39,720 --> 00:20:45,840
Speaker 3: people make decisions when we don't have perfect or complete information, which,

359
00:20:45,880 --> 00:20:48,800
Speaker 3: of course in life we rarely do. Right, So, you

360
00:20:48,840 --> 00:20:50,959
Speaker 3: know this is going to be relevant in understanding how

361
00:20:51,000 --> 00:20:55,240
Speaker 3: people behave in most real world situations. Most situations in

362
00:20:55,280 --> 00:20:57,879
Speaker 3: life are not a casino table game where you have

363
00:20:58,040 --> 00:20:59,520
Speaker 3: no odds of winning.

364
00:21:00,400 --> 00:21:02,600
Speaker 2: Or you know, or you know, you can easily imagine

365
00:21:02,600 --> 00:21:05,280
Speaker 2: to a situation where you're playing some sort of a game,

366
00:21:05,359 --> 00:21:07,520
Speaker 2: be it a tabletop role playing game or some sort

367
00:21:07,520 --> 00:21:10,000
Speaker 2: of like, you know, video game, and you're given a

368
00:21:10,080 --> 00:21:13,399
Speaker 2: choice between two items for your character. Right, you have

369
00:21:13,440 --> 00:21:15,600
Speaker 2: the stats of those items right there. Yeah, you know

370
00:21:15,640 --> 00:21:17,720
Speaker 2: what kind of enemies you go up against in the game.

371
00:21:17,760 --> 00:21:20,399
Speaker 2: Maybe you've played through the game before. You have something

372
00:21:20,440 --> 00:21:25,440
Speaker 2: at least approaching perfect knowledge of the simplified world, and

373
00:21:25,640 --> 00:21:28,399
Speaker 2: our real world is not so simplified and there's so

374
00:21:28,440 --> 00:21:29,480
Speaker 2: many variables to it.

375
00:21:29,640 --> 00:21:32,640
Speaker 3: That's exactly right. So the real world, we're constantly dealing

376
00:21:32,720 --> 00:21:36,199
Speaker 3: with uncertainty ambiguous situations where we don't have all the

377
00:21:36,240 --> 00:21:39,280
Speaker 3: information to judge what outcomes are most likely, and yet

378
00:21:39,320 --> 00:21:42,760
Speaker 3: somehow we navigate this world making decisions all the time.

379
00:21:43,520 --> 00:21:47,040
Speaker 3: The question being addressed here is how do people make

380
00:21:47,160 --> 00:21:50,560
Speaker 3: those decisions when they don't actually know what the relative

381
00:21:50,680 --> 00:21:55,680
Speaker 3: likelihood of different outcomes are. Subjective expected utility theory has

382
00:21:55,720 --> 00:21:59,080
Speaker 3: several rules, but a simplified version is that it says,

383
00:22:00,000 --> 00:22:03,840
Speaker 3: and then when people don't know the objective likelihood of

384
00:22:03,880 --> 00:22:08,760
Speaker 3: an outcome, they form beliefs about how likely that outcome is,

385
00:22:09,280 --> 00:22:12,720
Speaker 3: and then they act consistently with those beliefs in order

386
00:22:12,760 --> 00:22:17,080
Speaker 3: to maximize their personal benefit. I came across a passage

387
00:22:17,119 --> 00:22:20,480
Speaker 3: by the American economist Frank Knight which I think summarizes

388
00:22:20,480 --> 00:22:24,719
Speaker 3: this idea well. Night writes, quote, we must observe at

389
00:22:24,760 --> 00:22:27,760
Speaker 3: the outset that when an individual instance i e. A

390
00:22:27,840 --> 00:22:31,680
Speaker 3: one time event only is at issue, there is no

391
00:22:31,760 --> 00:22:36,360
Speaker 3: difference for conduct between a measurable risk and an unmeasurable uncertainty.

392
00:22:36,800 --> 00:22:40,200
Speaker 3: The individual, as already observed, throws his estimate of the

393
00:22:40,320 --> 00:22:43,760
Speaker 3: value of an opinion into the probability form of a

394
00:22:44,000 --> 00:22:48,240
Speaker 3: successes in B trials a slash B being a proper fraction,

395
00:22:48,720 --> 00:22:53,240
Speaker 3: and feels toward it as toward any other probability situation.

396
00:22:53,880 --> 00:22:57,880
Speaker 3: So we essentially we form a belief for a feeling

397
00:22:58,160 --> 00:23:01,560
Speaker 3: about how likely something is, and then we act as

398
00:23:01,560 --> 00:23:04,160
Speaker 3: if that was like you know that we know the

399
00:23:04,200 --> 00:23:06,720
Speaker 3: odds of a coin flip coming up heads are one

400
00:23:06,760 --> 00:23:09,400
Speaker 3: in two, whether or not we're actually right about those

401
00:23:09,400 --> 00:23:14,960
Speaker 3: feelings or beliefs. Elsberg used the principle of ambiguity aversion

402
00:23:15,040 --> 00:23:20,000
Speaker 3: to say, actually, no, people do not always behave as

403
00:23:20,040 --> 00:23:23,240
Speaker 3: if they have consistent beliefs about what is more or

404
00:23:23,320 --> 00:23:26,639
Speaker 3: less likely to happen. And you can prove this because

405
00:23:26,680 --> 00:23:30,960
Speaker 3: they make bets that would imply self contradictory beliefs in

406
00:23:31,000 --> 00:23:35,480
Speaker 3: the same situation, meaning they either don't actually have or

407
00:23:35,600 --> 00:23:39,919
Speaker 3: act on consistent beliefs about a situation, or they don't

408
00:23:39,920 --> 00:23:44,600
Speaker 3: always act to maximize their own monetary benefit. Elsberg's point

409
00:23:44,640 --> 00:23:50,680
Speaker 3: here was that we can have obscure previously unacknowledged motivations

410
00:23:50,880 --> 00:23:55,159
Speaker 3: like the desire to avoid dealing with ambiguity, and that

411
00:23:55,280 --> 00:23:59,960
Speaker 3: particular desire is so strong in some cases that it overrides.

412
00:24:00,000 --> 00:24:03,960
Speaker 3: It's our consistent thinking and a simple dollar value understanding

413
00:24:04,280 --> 00:24:08,120
Speaker 3: of rational self interest. And by the way, the technical

414
00:24:08,200 --> 00:24:11,120
Speaker 3: name of the savage axiom that Elsberg was attacking here

415
00:24:11,240 --> 00:24:15,960
Speaker 3: was it's called the sure thing principle. Another thing to

416
00:24:16,040 --> 00:24:19,320
Speaker 3: understand about Elsberg's paper was that he was arguing that

417
00:24:19,680 --> 00:24:24,320
Speaker 3: we need to make a stronger distinction between two different concepts.

418
00:24:24,800 --> 00:24:30,119
Speaker 3: Those are risk and ambiguity. Elsberg gives the definition that

419
00:24:30,440 --> 00:24:33,280
Speaker 3: risk is when we have a stake in an outcome

420
00:24:33,760 --> 00:24:36,879
Speaker 3: and we know the theoretical likelihood of that outcome. So

421
00:24:37,000 --> 00:24:39,560
Speaker 3: this is like the casino games or a coin toss.

422
00:24:39,600 --> 00:24:41,600
Speaker 3: You know, a fair coin toss is fifty to fifty.

423
00:24:41,680 --> 00:24:45,280
Speaker 3: We know what the outcome likelihood is, and we can

424
00:24:45,320 --> 00:24:48,080
Speaker 3: take a risk. A fair die roll is one in

425
00:24:48,240 --> 00:24:51,960
Speaker 3: six to get a particular number. Ambiguity is when we

426
00:24:52,160 --> 00:24:55,680
Speaker 3: don't have an objective way of knowing what the probability

427
00:24:55,680 --> 00:24:59,679
Speaker 3: of our desired outcome is. To Elsberg, ambiguity was quote

428
00:24:59,760 --> 00:25:03,280
Speaker 3: the the nature of one's information concerning the relative likelihood

429
00:25:03,320 --> 00:25:07,480
Speaker 3: of events, a quality depending on the amount, type, reliability,

430
00:25:07,520 --> 00:25:11,480
Speaker 3: and unanimity of information, giving rise to one's degree of

431
00:25:11,560 --> 00:25:16,040
Speaker 3: confidence in an estimation of relative likelihoods. And so it's

432
00:25:16,080 --> 00:25:19,080
Speaker 3: worth noting that you can have confidence or even what

433
00:25:19,119 --> 00:25:22,560
Speaker 3: feels like certainty, or act as if you have confidence

434
00:25:22,600 --> 00:25:25,760
Speaker 3: or certainty without actually being correct. You can go through

435
00:25:25,800 --> 00:25:30,720
Speaker 3: life having high confidence in objectively low probability things. One

436
00:25:30,760 --> 00:25:34,879
Speaker 3: of the actually really useful things about SEU theories, the

437
00:25:34,920 --> 00:25:39,280
Speaker 3: subjective expected utility theory, was that it made things simple

438
00:25:39,440 --> 00:25:43,920
Speaker 3: and easier to understand because it treated risk and ambiguity

439
00:25:44,200 --> 00:25:47,680
Speaker 3: the same. So a person making bets on a die

440
00:25:47,800 --> 00:25:50,880
Speaker 3: roll will know that the probability of rolling a specific

441
00:25:50,960 --> 00:25:53,719
Speaker 3: number is one in six, and they will behave accordingly.

442
00:25:54,400 --> 00:25:56,640
Speaker 3: And a person who has a stake in an outcome

443
00:25:56,720 --> 00:26:02,360
Speaker 3: with an unknown probability will, according to Savage's axioms, mentally

444
00:26:02,560 --> 00:26:08,199
Speaker 3: even subconsciously form an internal belief about the probability of

445
00:26:08,200 --> 00:26:12,679
Speaker 3: that outcome. So example is you imagine there is a

446
00:26:12,720 --> 00:26:16,119
Speaker 3: one in four chance it's going to reign today. That

447
00:26:16,240 --> 00:26:18,399
Speaker 3: might not have anything to do with reality, but you

448
00:26:18,560 --> 00:26:22,119
Speaker 3: just take that belief on and you behave as if

449
00:26:22,240 --> 00:26:24,840
Speaker 3: that is that you make all of your decisions based

450
00:26:24,880 --> 00:26:27,679
Speaker 3: on that belief. They will act according to that belief

451
00:26:28,040 --> 00:26:29,800
Speaker 3: the same way they would act to knowing that a

452
00:26:29,840 --> 00:26:33,760
Speaker 3: coin flip is fifty to fifty, and Elsberg argued, no,

453
00:26:34,160 --> 00:26:37,439
Speaker 3: that is not how we make decisions. SEU theory with

454
00:26:37,520 --> 00:26:42,840
Speaker 3: savagees axioms failed to account for other complexities in human behavior,

455
00:26:42,920 --> 00:26:47,280
Speaker 3: for example, our deep desire to get away from ambiguity,

456
00:26:47,520 --> 00:26:50,880
Speaker 3: Like we don't like ambiguity so strongly we will make

457
00:26:51,000 --> 00:26:55,080
Speaker 3: decisions that imply self contradictory beliefs just to avoid it. Now,

458
00:26:55,080 --> 00:26:57,399
Speaker 3: there have been a lot of arguments over the years

459
00:26:57,640 --> 00:27:03,240
Speaker 3: about how to interpret ambiguity aversion. Is it actually rational

460
00:27:03,560 --> 00:27:07,400
Speaker 3: in ways that were not previously recognized, or is it

461
00:27:07,520 --> 00:27:11,480
Speaker 3: more like some kind of systematic error or cognitive bias.

462
00:27:12,240 --> 00:27:14,600
Speaker 3: We might come back and revisit that debate in a

463
00:27:14,680 --> 00:27:18,000
Speaker 3: subsequent part of this series, but we've got a couple

464
00:27:18,080 --> 00:27:21,320
Speaker 3: other things we wanted to talk about today. First, Rob,

465
00:27:21,359 --> 00:27:23,520
Speaker 3: I know you have a really interesting place you want

466
00:27:23,560 --> 00:27:25,280
Speaker 3: to take this, but first I wanted to mention the

467
00:27:25,320 --> 00:27:30,359
Speaker 3: idea just of experimental confirmation, because one important thing to

468
00:27:30,440 --> 00:27:34,000
Speaker 3: understand is that Ellsberg's original paper did not have an

469
00:27:34,040 --> 00:27:37,480
Speaker 3: experimental component. It was a thought experiment, and he realized

470
00:27:37,480 --> 00:27:40,840
Speaker 3: the phenomenon needed to be tested with real human subjects.

471
00:27:41,359 --> 00:27:44,120
Speaker 3: In the decades since his paper, it has been tested many,

472
00:27:44,160 --> 00:27:48,800
Speaker 3: many times, many different ways, and generally it has proved robust.

473
00:27:49,560 --> 00:27:54,639
Speaker 3: There are some nuances and exceptions, but generally people behave

474
00:27:54,760 --> 00:27:58,119
Speaker 3: the way Elsberg predicted. They would a majority of people

475
00:27:58,440 --> 00:28:02,320
Speaker 3: structure their choices to void ambiguity, even when there's no

476
00:28:02,560 --> 00:28:06,400
Speaker 3: clear objective benefit to doing so, and they even make

477
00:28:06,440 --> 00:28:10,199
Speaker 3: bets that would imply against self contradictory beliefs. According to

478
00:28:10,320 --> 00:28:13,439
Speaker 3: SEU theory, if those bets can get them out of

479
00:28:13,520 --> 00:28:16,720
Speaker 3: dealing with ambiguous probabilities, so I don't have to mess

480
00:28:16,720 --> 00:28:22,359
Speaker 3: with that. Regarding those experiments, my main source here is

481
00:28:22,400 --> 00:28:27,920
Speaker 3: again that survey article by Machina and Siniskalchi. Again that's

482
00:28:28,480 --> 00:28:32,679
Speaker 3: ambiguity and ambiguity of version from twenty fourteen. The authors

483
00:28:32,680 --> 00:28:36,000
Speaker 3: here list a whole bunch of experiments where people carried

484
00:28:36,000 --> 00:28:38,240
Speaker 3: out versions of like the three color urn or this

485
00:28:38,360 --> 00:28:42,800
Speaker 3: other experiment that Delsberg described, the two urn experiment. They

486
00:28:42,880 --> 00:28:45,920
Speaker 3: list Felner in nineteen sixty one, which found that people

487
00:28:45,920 --> 00:28:49,360
Speaker 3: would rather take fifty to fifty odds than unknown odds,

488
00:28:49,400 --> 00:28:53,040
Speaker 3: which could be better or worse. They also cite Becker

489
00:28:53,080 --> 00:28:56,760
Speaker 3: and Brownson sixty four, m. Kremen sixty eight, Slovak and

490
00:28:56,840 --> 00:29:00,880
Speaker 3: Taverski from nineteen seventy four, Curly and Yate from eighty nine,

491
00:29:00,960 --> 00:29:05,800
Speaker 3: among others, all generally finding confirmation of Elsberg's predictions about

492
00:29:05,840 --> 00:29:11,000
Speaker 3: ambiguity aversion quote. Although most of these experiments used students

493
00:29:11,080 --> 00:29:14,880
Speaker 3: as subjects. Researchers such as McCrimmon sixty five, Hogarth and

494
00:29:15,360 --> 00:29:20,200
Speaker 3: Kunruther eighty nine, Einhorn and Hogarth eighty six, Viscusi and

495
00:29:20,280 --> 00:29:24,160
Speaker 3: Chessen ninety nine, hoe Keller and Celtica two thousand and two,

496
00:29:24,480 --> 00:29:28,240
Speaker 3: and Mafioletti and Sentori two thousand and five have examined

497
00:29:28,240 --> 00:29:33,480
Speaker 3: the ambiguity preferences of business owners, trade union leaders, actuaries, managers,

498
00:29:33,520 --> 00:29:37,240
Speaker 3: and executives with the same overall findings, so it appears

499
00:29:37,320 --> 00:29:42,400
Speaker 3: quite robust. The authors here also report some interesting findings

500
00:29:42,480 --> 00:29:46,680
Speaker 3: of McCrimmon and Larsen from seventy nine which they looked

501
00:29:46,720 --> 00:29:49,000
Speaker 3: into it, and they found that while the majority of

502
00:29:49,040 --> 00:29:53,760
Speaker 3: people did try to avoid ambiguity, some minority of people

503
00:29:54,000 --> 00:29:57,440
Speaker 3: did not and even just chose to embrace it. They

504
00:29:57,480 --> 00:30:01,360
Speaker 3: also found, as you might expect, that our relative tolerance

505
00:30:01,400 --> 00:30:04,880
Speaker 3: for ambiguity went up or down depending on how good

506
00:30:05,000 --> 00:30:09,880
Speaker 3: the known odds were in the known odds bet condition. So,

507
00:30:10,000 --> 00:30:12,240
Speaker 3: for example, if you do this three color earn experiment,

508
00:30:12,520 --> 00:30:14,680
Speaker 3: if you take the number of red balls that's the

509
00:30:14,680 --> 00:30:16,520
Speaker 3: ones where you know how many there are, If you

510
00:30:16,560 --> 00:30:19,000
Speaker 3: take that down to zero or even just down to

511
00:30:19,080 --> 00:30:22,400
Speaker 3: five or ten, most people will take the gamble with

512
00:30:22,480 --> 00:30:25,719
Speaker 3: the unknown odds on yellow and black because the odds

513
00:30:25,760 --> 00:30:28,680
Speaker 3: on red are clearly very bad. If you make the

514
00:30:28,720 --> 00:30:32,240
Speaker 3: odds on red really good, even more people will stick

515
00:30:32,280 --> 00:30:35,400
Speaker 3: with the no nods, and ambiguity of version becomes even stronger.

516
00:30:35,440 --> 00:30:38,280
Speaker 3: So that shouldn't be all that surprising. But yeah, in

517
00:30:38,600 --> 00:30:41,840
Speaker 3: the middle category, where the odds on red are somewhere

518
00:30:41,880 --> 00:30:44,520
Speaker 3: in between good and bad, you're just kind of like,

519
00:30:44,760 --> 00:30:47,560
Speaker 3: I don't know, But still people tend to go for

520
00:30:47,760 --> 00:30:51,720
Speaker 3: the known odds rather than the unknown odds. However, the

521
00:30:51,800 --> 00:30:57,960
Speaker 3: authors also found that even when the idea of subjective

522
00:30:57,960 --> 00:31:01,600
Speaker 3: expected utility and choice consistent see was explained to the

523
00:31:01,640 --> 00:31:05,320
Speaker 3: people doing this experiment explained to the subjects, or even

524
00:31:05,320 --> 00:31:09,680
Speaker 3: in cases where the known odds were bad, some subjects

525
00:31:09,800 --> 00:31:13,920
Speaker 3: just stubbornly stuck to red and avoided betting on any

526
00:31:14,000 --> 00:31:19,120
Speaker 3: ambiguous conditions. So most of us seem to dislike ambiguity

527
00:31:19,720 --> 00:31:22,400
Speaker 3: in the majority of cases, and some of us just

528
00:31:22,440 --> 00:31:35,560
Speaker 3: don't like it at all and will not tolerate it. Now,

529
00:31:35,600 --> 00:31:38,959
Speaker 3: as is often the case with stuff and decision theory

530
00:31:39,080 --> 00:31:42,600
Speaker 3: and economics and stuff, you start off playing these little

531
00:31:42,640 --> 00:31:45,040
Speaker 3: like gambling games with like balls in a jar or

532
00:31:45,040 --> 00:31:47,040
Speaker 3: something like that, and so it seems like, well, how

533
00:31:47,080 --> 00:31:50,200
Speaker 3: consequential could this really be? But actually, I think you

534
00:31:50,240 --> 00:31:54,520
Speaker 3: can take the idea of ambiguity aversion to number one.

535
00:31:54,600 --> 00:31:58,720
Speaker 3: It has been applied to very consequential situations in real life,

536
00:31:58,800 --> 00:32:03,360
Speaker 3: and thus understanding it can have major consequences not only

537
00:32:03,400 --> 00:32:06,400
Speaker 3: for people's individual lives, but for world events.

538
00:32:07,200 --> 00:32:10,760
Speaker 2: That's right, And that brings us back to Ellsberg, who, again,

539
00:32:10,840 --> 00:32:15,880
Speaker 2: and in addition to crafting this highly influential paper, was

540
00:32:15,920 --> 00:32:20,960
Speaker 2: also a highly vocal critic of nuclear weapons and again

541
00:32:21,000 --> 00:32:26,280
Speaker 2: the rhetoric surrounding nuclear weapons. And I was reading a

542
00:32:26,280 --> 00:32:28,920
Speaker 2: good bit from his book The Doomsday Machine, which came

543
00:32:28,920 --> 00:32:31,440
Speaker 2: out in twenty twelve, which is quite a good read

544
00:32:31,480 --> 00:32:33,480
Speaker 2: if you want to much deeper dive into his thoughts

545
00:32:33,560 --> 00:32:37,640
Speaker 2: on all of this. But you know, in that book

546
00:32:37,960 --> 00:32:41,960
Speaker 2: the subject of ambiguous data comes up numerous times how

547
00:32:42,040 --> 00:32:46,800
Speaker 2: will a nuclear power react in response to ambiguous data

548
00:32:47,160 --> 00:32:51,600
Speaker 2: about a potential incoming attack? And additionally, it talks about

549
00:32:51,640 --> 00:32:57,600
Speaker 2: the purported advantage of ambiguity in intense slash response, namely

550
00:32:57,640 --> 00:33:01,560
Speaker 2: in the case of mad Man theory. This is very

551
00:33:01,560 --> 00:33:06,000
Speaker 2: closely associated with Richard Nixon, but also tied to various

552
00:33:06,040 --> 00:33:10,400
Speaker 2: other figures domestic and international, including the current US president.

553
00:33:10,720 --> 00:33:16,280
Speaker 2: A kind of supposed perceived madness or volatility that would

554
00:33:16,400 --> 00:33:20,400
Speaker 2: lead adversaries to second guests any plans to move against them.

555
00:33:20,840 --> 00:33:23,440
Speaker 3: Right, and we're going to get to some major criticisms

556
00:33:23,480 --> 00:33:26,120
Speaker 3: of the mad Man theory strategy in a bit, But

557
00:33:26,200 --> 00:33:29,800
Speaker 3: the idea of it is that you can leverage a

558
00:33:29,920 --> 00:33:35,320
Speaker 3: reputation for unpredictability to your advantage in negotiations.

559
00:33:35,960 --> 00:33:38,080
Speaker 2: That's right. You know, this is very much that nobody

560
00:33:38,120 --> 00:33:41,640
Speaker 2: knows what I'm going to do approach. That's the message

561
00:33:41,880 --> 00:33:44,479
Speaker 2: you put out there. And the idea part of the

562
00:33:44,520 --> 00:33:47,720
Speaker 2: idea here is that it makes logically empty or extreme

563
00:33:47,880 --> 00:33:51,560
Speaker 2: threats seem more possible. Now, mad Man theory was you.

564
00:33:51,560 --> 00:33:54,120
Speaker 2: Appointment is a term that Nixon even used himself when

565
00:33:54,160 --> 00:33:57,400
Speaker 2: talking about this approach. It was outlined by Daniel Elsberg

566
00:33:57,440 --> 00:33:59,880
Speaker 2: and Thomas C. Showing as well, but the basic con

567
00:34:00,720 --> 00:34:04,480
Speaker 2: pretending to be erratic, irrational, or unhinged to influence the

568
00:34:04,480 --> 00:34:07,040
Speaker 2: other side of a bargaining table. This goes back a

569
00:34:07,080 --> 00:34:11,040
Speaker 2: long ways. Machiavelli wrote about it in fifteen seventeen, and

570
00:34:11,080 --> 00:34:14,239
Speaker 2: it's probably as old as the first prehistoric human to

571
00:34:14,280 --> 00:34:17,640
Speaker 2: like go around hitting trees randomly with a club, you know,

572
00:34:17,719 --> 00:34:19,839
Speaker 2: and say, who knows what tharg is going to do?

573
00:34:20,400 --> 00:34:22,759
Speaker 2: Why would you mess with throg on this? We?

574
00:34:22,880 --> 00:34:24,840
Speaker 3: Yeah, we better just give throarc what he wants.

575
00:34:25,160 --> 00:34:29,240
Speaker 2: Yeah, So again, the Elsberg paradox highlights a core flaw

576
00:34:29,320 --> 00:34:33,360
Speaker 2: in human decision making, the aversion to ambiguity. The madman

577
00:34:33,520 --> 00:34:36,759
Speaker 2: theory as a political strategy seeks to exploit this by

578
00:34:36,800 --> 00:34:41,799
Speaker 2: cultivating a reputation of rationality, creating an ambiguous threat. The

579
00:34:41,880 --> 00:34:45,160
Speaker 2: adversary and their aversion to this ambiguity and the unknown

580
00:34:45,239 --> 00:34:50,160
Speaker 2: risk of an unpredictable response may be coerced into backing down.

581
00:34:50,239 --> 00:34:52,000
Speaker 2: Or at least this is the idea, this is the

582
00:34:52,680 --> 00:34:54,800
Speaker 2: logic behind the approach. Yeah.

583
00:34:55,040 --> 00:34:57,880
Speaker 3: The classic example of mad men theory in practice in

584
00:34:57,920 --> 00:35:02,560
Speaker 3: real life is that the Nixon administration could, for example,

585
00:35:02,840 --> 00:35:07,000
Speaker 3: have Kissinger call up his counterparts in the Kremlin and say, look,

586
00:35:07,239 --> 00:35:09,239
Speaker 3: you know, we're all trying to calm Nixon down, but

587
00:35:09,280 --> 00:35:12,959
Speaker 3: he's boiling mad at you. He's borderline crazy. You've got

588
00:35:12,960 --> 00:35:14,759
Speaker 3: to do X, Y and Z. You know, you've got

589
00:35:14,760 --> 00:35:16,879
Speaker 3: to give him these concessions he wants, or we don't

590
00:35:16,920 --> 00:35:18,560
Speaker 3: know if we'll be able to control him.

591
00:35:19,000 --> 00:35:20,960
Speaker 2: Yeah. Kind of a good cut, bad cotton thing too,

592
00:35:21,040 --> 00:35:21,719
Speaker 2: I imagine.

593
00:35:21,840 --> 00:35:25,799
Speaker 3: Yeah, And this was a way of making I think

594
00:35:25,840 --> 00:35:27,399
Speaker 3: a big part of it was it was a way

595
00:35:27,440 --> 00:35:31,120
Speaker 3: of making the nuclear deterrent feel once again like it

596
00:35:31,200 --> 00:35:35,440
Speaker 3: had force in matters other than just deterring a first strike.

597
00:35:36,440 --> 00:35:39,880
Speaker 3: Because the problem with using nuclear weapons as leverage and

598
00:35:39,960 --> 00:35:44,759
Speaker 3: negotiation between two nuclear armed powers is obviously, if the

599
00:35:44,760 --> 00:35:48,480
Speaker 3: weapons get used at all, everyone loses. There is not

600
00:35:48,600 --> 00:35:52,280
Speaker 3: a winner in mutually sured destruction. So if the White

601
00:35:52,280 --> 00:35:54,960
Speaker 3: House was trying to get the Kremlin to do something

602
00:35:55,040 --> 00:35:58,160
Speaker 3: like I'll nuke if you don't do it, that was

603
00:35:58,200 --> 00:36:01,560
Speaker 3: not a credible threat because bo both parties knew that

604
00:36:01,600 --> 00:36:04,760
Speaker 3: the other party knows that war would be the destruction

605
00:36:04,840 --> 00:36:08,920
Speaker 3: of them both. So no rational actor would ever strike first.

606
00:36:09,400 --> 00:36:12,480
Speaker 3: Between rational actors, I'll knew Q if you don't do

607
00:36:12,719 --> 00:36:16,600
Speaker 3: XYZ is an empty threat. The goal of madman theory

608
00:36:17,080 --> 00:36:21,000
Speaker 3: is to gain leverage in negotiations by introducing some amount

609
00:36:21,080 --> 00:36:25,400
Speaker 3: of fear that your counterparty may not be a rational actor.

610
00:36:25,719 --> 00:36:29,279
Speaker 3: He might be crazy enough to do it, we don't know. Therefore,

611
00:36:29,560 --> 00:36:33,960
Speaker 3: the madman strategy makes direct use of the opponent's ambiguity

612
00:36:34,040 --> 00:36:37,640
Speaker 3: version as a tactic to get leverage over them. The

613
00:36:37,680 --> 00:36:40,880
Speaker 3: madman player in this game is betting that the fear

614
00:36:40,960 --> 00:36:45,920
Speaker 3: and uncertainty about how to handle their unpredictability will cause

615
00:36:45,960 --> 00:36:49,520
Speaker 3: their counterparty to make concessions they might not make otherwise,

616
00:36:49,680 --> 00:36:52,319
Speaker 3: Like you end up thinking, well, I'll just take a

617
00:36:52,360 --> 00:36:56,120
Speaker 3: bad deal rather than wander into territory with unclear levels of.

618
00:36:56,200 --> 00:37:00,520
Speaker 2: Risk, right right, and not to derail the pointer, But

619
00:37:00,560 --> 00:37:04,120
Speaker 2: I want to bring it back to Ellsberg's other criticisms

620
00:37:04,160 --> 00:37:07,399
Speaker 2: of nuclear weapons and the rhetoric surrounding nuclear weapons, that being,

621
00:37:07,560 --> 00:37:10,640
Speaker 2: there's all the there's all these other uncertainty factors and

622
00:37:10,760 --> 00:37:12,960
Speaker 2: ambiguous data that would be coming in about what the

623
00:37:13,080 --> 00:37:17,040
Speaker 2: enemy is or is not doing, and therefore any given

624
00:37:17,080 --> 00:37:20,319
Speaker 2: situation like this is going to be just rot with

625
00:37:20,320 --> 00:37:24,279
Speaker 2: with potential ambiguity, potential missteps and potential mistakes, So like

626
00:37:24,360 --> 00:37:26,759
Speaker 2: if it were only as easy as we're laying it

627
00:37:26,800 --> 00:37:29,760
Speaker 2: out here, it would almost be a different situation totally.

628
00:37:29,800 --> 00:37:34,120
Speaker 3: I mean Ellsberg's point about the all of the uncertainty

629
00:37:34,360 --> 00:37:38,240
Speaker 3: in the actual command and control of you know, nuclear

630
00:37:38,239 --> 00:37:43,959
Speaker 3: strategy like that's that that introduces real levels of existential

631
00:37:44,040 --> 00:37:47,839
Speaker 3: risk into playing games like this. There are a lot

632
00:37:47,920 --> 00:37:50,680
Speaker 3: of reasonable criticisms of mad Man theory, but I think

633
00:37:50,680 --> 00:37:52,839
Speaker 3: a major one, at least from my point of view,

634
00:37:52,920 --> 00:37:57,120
Speaker 3: is that at best it is a strategy for short

635
00:37:57,239 --> 00:38:01,040
Speaker 3: term gain at the expense of long term stability and

636
00:38:01,120 --> 00:38:05,719
Speaker 3: negotiating power, because one way or another it undercuts your

637
00:38:05,719 --> 00:38:09,080
Speaker 3: credibility and your ability to be seen as an honest

638
00:38:09,120 --> 00:38:13,600
Speaker 3: and reliable broker in future negotiations, either because you come

639
00:38:13,680 --> 00:38:17,160
Speaker 3: to be seen as actually irrational, in which case people

640
00:38:17,200 --> 00:38:18,960
Speaker 3: don't want to deal with you, and they you know,

641
00:38:19,040 --> 00:38:21,799
Speaker 3: they will find what you know. They may be they

642
00:38:21,800 --> 00:38:24,480
Speaker 3: may be motivated to do something bad to strike first

643
00:38:24,480 --> 00:38:26,840
Speaker 3: to you because you're too dangerous to be let loose,

644
00:38:26,960 --> 00:38:29,440
Speaker 3: or they may be, you know, they may try to

645
00:38:29,480 --> 00:38:32,440
Speaker 3: find ways to cut you out of negotiations, so that

646
00:38:32,440 --> 00:38:35,120
Speaker 3: could be one consequence, or just because you are revealed

647
00:38:35,160 --> 00:38:38,680
Speaker 3: to have been bluffing, in which case you lose your credibility.

648
00:38:38,520 --> 00:38:41,000
Speaker 2: Right right, and yeah, this is this is all valid.

649
00:38:41,520 --> 00:38:43,759
Speaker 2: I was looking at a couple of papers on mad

650
00:38:43,800 --> 00:38:47,880
Speaker 2: men theory and mad man theory, like ambiguity of version itself,

651
00:38:47,920 --> 00:38:50,520
Speaker 2: is something that has been written about a lot there.

652
00:38:50,640 --> 00:38:53,279
Speaker 2: There are no shortage of papers out there, but I

653
00:38:53,280 --> 00:38:56,040
Speaker 2: looked at just a couple. I looked at one titled

654
00:38:56,160 --> 00:38:59,240
Speaker 2: crazy like a Fox? Are leaders with reputations for madness

655
00:38:59,239 --> 00:39:04,480
Speaker 2: more successful international coercion? By Roseanne W McManus. This came

656
00:39:04,520 --> 00:39:07,840
Speaker 2: out in twenty nineteen from Cambridge University Press, and in

657
00:39:07,880 --> 00:39:10,880
Speaker 2: this the author points out that a reputation for madness

658
00:39:11,120 --> 00:39:15,160
Speaker 2: would seem to be more often harmful than helpful in

659
00:39:15,160 --> 00:39:20,359
Speaker 2: international coercion, So it undercuts the leader's ability to make

660
00:39:20,440 --> 00:39:24,920
Speaker 2: believable peace commitments, treaties and so forth, which and it

661
00:39:25,000 --> 00:39:27,160
Speaker 2: really almost feels absurd that we need to stress this.

662
00:39:27,719 --> 00:39:31,719
Speaker 2: These are all vital for peaceful relations between nations. No

663
00:39:31,760 --> 00:39:34,480
Speaker 2: one knows what I'll do easily slides into. No one

664
00:39:34,560 --> 00:39:37,799
Speaker 2: knows what I'll honor, No one knows what treaties I

665
00:39:37,880 --> 00:39:41,880
Speaker 2: would well I actually stand by, and so forth. You know,

666
00:39:41,920 --> 00:39:43,640
Speaker 2: to bring this back to Dungeons and Dragons, I think

667
00:39:43,640 --> 00:39:45,680
Speaker 2: any dn D player worth your salt knows that while

668
00:39:45,719 --> 00:39:48,120
Speaker 2: you might well enter into a pact with a lawful

669
00:39:48,200 --> 00:39:51,160
Speaker 2: evil devil, you never enter into a pack with a

670
00:39:51,239 --> 00:39:54,480
Speaker 2: chaotic evil demon because the lords of the nine hell

671
00:39:54,520 --> 00:39:56,400
Speaker 2: are going to stick to their letter of the contract,

672
00:39:56,400 --> 00:39:59,400
Speaker 2: if not the spirit of the contract. But demon lords

673
00:39:59,440 --> 00:40:02,560
Speaker 2: such as the mcgorgan will honor nothing, and there is

674
00:40:02,719 --> 00:40:06,600
Speaker 2: no coherence even within themselves. They're just pure ambiguity, and

675
00:40:06,640 --> 00:40:09,680
Speaker 2: you can't strike any sort of deal because they won't

676
00:40:09,719 --> 00:40:12,359
Speaker 2: stand by it no matter what. Whereas the devil's being

677
00:40:12,440 --> 00:40:15,960
Speaker 2: lawful evil, they may look for that wiggle room, they

678
00:40:16,000 --> 00:40:19,600
Speaker 2: may find ways to, you know, to avoid the spirit

679
00:40:19,719 --> 00:40:21,720
Speaker 2: of the deal, but they're still bound to the letter.

680
00:40:22,040 --> 00:40:24,399
Speaker 3: Right. So if you think you discover that your your

681
00:40:24,440 --> 00:40:27,440
Speaker 3: counterparty is actually just chaotic evil, all you can do

682
00:40:27,520 --> 00:40:29,960
Speaker 3: is roll for initiative like there's no making a.

683
00:40:29,920 --> 00:40:33,080
Speaker 2: Deal yeah, and then you slide into chaos, pure chaos.

684
00:40:34,160 --> 00:40:37,120
Speaker 2: McManus back to her paper, though, she found that mad

685
00:40:37,160 --> 00:40:40,960
Speaker 2: men theory may be helpful in crisis bargaining, but only

686
00:40:41,080 --> 00:40:46,240
Speaker 2: under certain conditions, namely when employed by military week leaders.

687
00:40:46,640 --> 00:40:49,520
Speaker 2: So we're not talking about a true superpower here they

688
00:40:49,560 --> 00:40:53,759
Speaker 2: can engage in like true mutually assured destruction. Rather, we

689
00:40:53,800 --> 00:40:56,440
Speaker 2: would be dealing with the state that could do a

690
00:40:56,440 --> 00:41:00,640
Speaker 2: lot of damage, but in striking out with just thoroughly

691
00:41:00,640 --> 00:41:03,759
Speaker 2: destroy themselves. So the idea here is that no one

692
00:41:03,800 --> 00:41:07,360
Speaker 2: would take this hypothetical nation or the leader of this

693
00:41:07,440 --> 00:41:12,759
Speaker 2: hypothetical nation seriously unless they presented an air of madness,

694
00:41:12,800 --> 00:41:16,160
Speaker 2: and then perhaps, hey, they might do it anyway if provoked.

695
00:41:16,200 --> 00:41:18,480
Speaker 2: They they you know, they're they're unhinged, and thus it

696
00:41:18,520 --> 00:41:23,040
Speaker 2: can be a form of asymmetric leverage. So we're talking

697
00:41:23,239 --> 00:41:28,000
Speaker 2: low probability but high consequence. And MacManus also points out

698
00:41:28,000 --> 00:41:29,560
Speaker 2: that there would also seem to be more of an

699
00:41:29,560 --> 00:41:33,200
Speaker 2: advantage if you had a mild reputation for madness rather

700
00:41:33,239 --> 00:41:35,480
Speaker 2: than an extreme one. You don't want to come off like,

701
00:41:35,880 --> 00:41:41,200
Speaker 2: you know, a complete da da you know, eye rolling maniac.

702
00:41:41,560 --> 00:41:43,720
Speaker 2: In one of these cases, the idea would be like, well,

703
00:41:44,000 --> 00:41:46,000
Speaker 2: you know, we can, we can still get this individual

704
00:41:46,080 --> 00:41:48,960
Speaker 2: to the negotiation table, but we just have to be

705
00:41:49,239 --> 00:41:53,160
Speaker 2: hyper sensitive and I guess again, kind of avoiding that

706
00:41:53,320 --> 00:41:57,840
Speaker 2: complete fall off into chaotic evil demonhood and instead dealing

707
00:41:57,920 --> 00:41:59,879
Speaker 2: with devils that can still be bound by some sort

708
00:41:59,880 --> 00:42:03,640
Speaker 2: of law. All right, I have a hypothetical example here.

709
00:42:03,760 --> 00:42:08,799
Speaker 2: So we have demigodic hero Hercules and we have his

710
00:42:09,000 --> 00:42:13,240
Speaker 2: mortal cousin Eurystheus, and they're grabbing lunch at the local

711
00:42:13,239 --> 00:42:18,239
Speaker 2: euro place. Great, So Eurystheus warns his cousin that if

712
00:42:18,239 --> 00:42:20,480
Speaker 2: he tries to steal any of his fries, he's going

713
00:42:20,520 --> 00:42:22,120
Speaker 2: to flip the table and all their food's going to

714
00:42:22,160 --> 00:42:24,480
Speaker 2: wind up on the floor, and he acts just really

715
00:42:24,520 --> 00:42:29,040
Speaker 2: sensitive and unhinged about the whole thing. Yes, Okay, Hercules

716
00:42:29,200 --> 00:42:33,000
Speaker 2: obviously knows that Eurystheus can't take him in a fight

717
00:42:33,120 --> 00:42:35,280
Speaker 2: and might not even be able to flip that table

718
00:42:35,320 --> 00:42:41,760
Speaker 2: over before Hercules stops him, and rationally, Eurystheus should realize

719
00:42:41,760 --> 00:42:44,080
Speaker 2: this and just let big Hurk have a few extra

720
00:42:44,120 --> 00:42:45,839
Speaker 2: fries if he wants, like that's what does it matter.

721
00:42:46,520 --> 00:42:50,280
Speaker 2: But from Hercules's standpoint, does he really want to risk

722
00:42:50,400 --> 00:42:53,280
Speaker 2: his lunch winding up on the floor. Maybe he should

723
00:42:53,320 --> 00:42:57,680
Speaker 2: just let Eurystheus keep all of his fries. Okay. The

724
00:42:57,800 --> 00:43:02,600
Speaker 2: downside of course here is that maybe Hercules just won't

725
00:43:02,640 --> 00:43:05,439
Speaker 2: invite you risk the us out for lunch next time,

726
00:43:05,440 --> 00:43:07,160
Speaker 2: and he's certainly not going to pick up the check.

727
00:43:07,719 --> 00:43:10,879
Speaker 2: And on top of all that, how are they going

728
00:43:10,920 --> 00:43:13,759
Speaker 2: to work together to defeat the hydra? Because, as you

729
00:43:13,880 --> 00:43:16,400
Speaker 2: remember from past episodes and from Greek mythology in general,

730
00:43:16,640 --> 00:43:18,920
Speaker 2: Hercules can't do that on his own. He has to

731
00:43:18,960 --> 00:43:21,479
Speaker 2: have his cousins help to burn the stomps after each

732
00:43:21,640 --> 00:43:23,160
Speaker 2: head of the hydras cut off.

733
00:43:23,080 --> 00:43:27,200
Speaker 3: Very good points, I mean, apart from any like otherwise

734
00:43:27,239 --> 00:43:31,400
Speaker 3: like moral considerations or honesty considerations about deploying something like

735
00:43:31,480 --> 00:43:33,959
Speaker 3: mad men theory. Just from a strategic point of view,

736
00:43:35,200 --> 00:43:38,680
Speaker 3: it seems like it is probably best for situations where

737
00:43:39,160 --> 00:43:42,520
Speaker 3: like your counterparty has to deal with you, it is

738
00:43:42,600 --> 00:43:46,759
Speaker 3: not optional for them, and you don't have to like

739
00:43:47,080 --> 00:43:50,480
Speaker 3: or respect each other or ever work together on anything,

740
00:43:51,640 --> 00:43:54,600
Speaker 3: and you don't care about long term goals. You are

741
00:43:54,680 --> 00:43:58,719
Speaker 3: only interested in this situation right now, in extracting a

742
00:43:58,760 --> 00:44:00,120
Speaker 3: short term advantage.

743
00:44:00,760 --> 00:44:04,160
Speaker 2: That's right, that's right. But some papers out there that

744
00:44:04,239 --> 00:44:06,640
Speaker 2: have crunched the numbers on all this and done some

745
00:44:07,080 --> 00:44:10,719
Speaker 2: experimentation questionnaires and so forth, do acknowledge that you know

746
00:44:10,719 --> 00:44:14,640
Speaker 2: that there are multiple dimensions to any of these situations. Namely,

747
00:44:15,000 --> 00:44:17,000
Speaker 2: and this is something I think everyone can relate to, Like,

748
00:44:17,040 --> 00:44:20,520
Speaker 2: you can have a leader that is engaging in an

749
00:44:20,520 --> 00:44:26,120
Speaker 2: international situation, but there is still the domestic view of

750
00:44:26,160 --> 00:44:29,839
Speaker 2: that situation. There's still the domestic response, the domestic relationship.

751
00:44:31,000 --> 00:44:32,839
Speaker 2: And so one of the papers that I was looking

752
00:44:32,880 --> 00:44:35,440
Speaker 2: at this one is titled mad Man or mad Genius.

753
00:44:35,440 --> 00:44:38,840
Speaker 2: The International Benefits and Domestic Costs of mad Man's Strategy

754
00:44:38,880 --> 00:44:42,840
Speaker 2: by Joshua A. Schwartz, published twenty twenty three in Security Studies,

755
00:44:43,640 --> 00:44:46,400
Speaker 2: and in this the author found that the Madman approach

756
00:44:46,560 --> 00:44:51,160
Speaker 2: can work in negotiations with foreign adversaries, but quote entails

757
00:44:51,200 --> 00:44:55,560
Speaker 2: significant domestic costs that potentially erode its efficacy. He also

758
00:44:55,640 --> 00:44:58,200
Speaker 2: points out that mad Man theory may simply not work

759
00:44:58,239 --> 00:45:01,880
Speaker 2: against major powers, because while it might make a threat

760
00:45:01,920 --> 00:45:05,480
Speaker 2: more credible, that doesn't necessarily make the threat more effective

761
00:45:06,040 --> 00:45:08,960
Speaker 2: and make the adversary willing to cave to those demands.

762
00:45:09,000 --> 00:45:12,280
Speaker 2: So he points out that this is perhaps why Nixon's

763
00:45:12,320 --> 00:45:15,400
Speaker 2: use of mad Man theory didn't work against the Soviets. Like, Okay,

764
00:45:15,440 --> 00:45:20,480
Speaker 2: it made Nixon's threat more credible, but did it actually

765
00:45:20,719 --> 00:45:22,799
Speaker 2: make it more effective? Did he actually get what he

766
00:45:22,880 --> 00:45:26,560
Speaker 2: wanted to out of these bluffs these threats, and then

767
00:45:26,600 --> 00:45:30,480
Speaker 2: you have the domestic side of things, with the hypothetical

768
00:45:30,640 --> 00:45:36,000
Speaker 2: leader's own citizens not loving the heightened stakes because you know,

769
00:45:36,120 --> 00:45:38,920
Speaker 2: obviously I mean, I say obviously, but this is one

770
00:45:38,920 --> 00:45:40,480
Speaker 2: of those things that we often have to be reminded

771
00:45:40,520 --> 00:45:43,680
Speaker 2: of in a case of mutually assured destruction, or even

772
00:45:43,719 --> 00:45:47,680
Speaker 2: a case of asymmetric exchange involving nuclear weapons or some

773
00:45:47,719 --> 00:45:50,280
Speaker 2: other kind of just you know, horrible weapon of mass destruction,

774
00:45:51,160 --> 00:45:54,000
Speaker 2: a leader is always bargaining with the lives of their

775
00:45:54,000 --> 00:45:56,719
Speaker 2: own citizens. You are the chips on the board in

776
00:45:56,760 --> 00:45:59,600
Speaker 2: a no one knows what I'll do wager. Yeah, And

777
00:45:59,600 --> 00:46:02,520
Speaker 2: there also notes that, okay, you know, this is going

778
00:46:02,600 --> 00:46:04,760
Speaker 2: to also differ depending on how much of a voice

779
00:46:04,760 --> 00:46:07,120
Speaker 2: the people have in a given nation versus how much

780
00:46:07,280 --> 00:46:10,799
Speaker 2: power the ruler has, and there's going to be less

781
00:46:10,840 --> 00:46:13,520
Speaker 2: backlash in places where the people have less of a

782
00:46:13,640 --> 00:46:15,800
Speaker 2: voice in the ruler has more absolute power.

783
00:46:16,000 --> 00:46:18,280
Speaker 3: That's right. But one thing I did want to clarify

784
00:46:18,520 --> 00:46:21,640
Speaker 3: is that the sense in which you are the chips

785
00:46:21,680 --> 00:46:23,680
Speaker 3: on the board and no one knows what I'll do

786
00:46:23,800 --> 00:46:28,879
Speaker 3: wager is true is not just limited to like the

787
00:46:28,920 --> 00:46:32,080
Speaker 3: worst possible scenario like nuclear warfare. I mean it's also

788
00:46:32,680 --> 00:46:35,640
Speaker 3: the case in say, like trade negotiations or something like that,

789
00:46:35,680 --> 00:46:39,120
Speaker 3: like your economic prospects are in a way, those are

790
00:46:39,160 --> 00:46:43,080
Speaker 3: the chips on the board, Like the citizens' fates and

791
00:46:43,160 --> 00:46:45,759
Speaker 3: futures are the things that are being negotiated with.

792
00:46:46,120 --> 00:46:51,160
Speaker 2: That's right. Yeah, in contemplating potential nuclear exchanges, everything is

793
00:46:51,200 --> 00:46:54,000
Speaker 2: a bit more stark and a bit more black and white.

794
00:46:54,040 --> 00:46:56,800
Speaker 2: But obviously, you know, any give who can think of

795
00:46:56,800 --> 00:46:59,640
Speaker 2: any number of scenarios where the steaks are are still

796
00:46:59,719 --> 00:47:01,600
Speaker 2: quite high for the individual.

797
00:47:01,800 --> 00:47:03,719
Speaker 3: So yes, it's not hard to see at all why

798
00:47:03,800 --> 00:47:07,839
Speaker 3: a citizenry can easily become upset and annoyed if they're

799
00:47:07,960 --> 00:47:11,080
Speaker 3: leader who's supposed to be representing their interests in negotiations,

800
00:47:11,160 --> 00:47:15,879
Speaker 3: is acting unreliable and unpredictable. Like the leader may think

801
00:47:15,920 --> 00:47:17,960
Speaker 3: that they can get good gains out of that in

802
00:47:18,000 --> 00:47:21,120
Speaker 3: the short term, but the citizens are probably thinking about

803
00:47:21,120 --> 00:47:23,680
Speaker 3: a lot of these even if not you know, thinking

804
00:47:23,719 --> 00:47:26,840
Speaker 3: specifically about them, just thinking intuitively about it. Seems like

805
00:47:26,840 --> 00:47:28,239
Speaker 3: there's a lot of downside to this.

806
00:47:28,920 --> 00:47:32,400
Speaker 2: Yeah, and again coming back to what we mentioned earlier,

807
00:47:32,440 --> 00:47:35,920
Speaker 2: the idea too that by constant, if you swing around

808
00:47:36,560 --> 00:47:39,000
Speaker 2: the sword of mad Man theory too much, then it

809
00:47:39,040 --> 00:47:42,800
Speaker 2: makes it more difficult to engage in the other highly

810
00:47:42,840 --> 00:47:47,719
Speaker 2: important tools of state craft, peace treaties and agreements and

811
00:47:47,760 --> 00:47:50,520
Speaker 2: so forth. So it's like you can only swing that

812
00:47:50,560 --> 00:47:53,600
Speaker 2: sword around for so long, and then you have to

813
00:47:53,600 --> 00:47:57,120
Speaker 2: be able to engage in these other acts as well. Again,

814
00:47:57,360 --> 00:48:01,239
Speaker 2: short term gain with mad Man theory, not so in

815
00:48:01,280 --> 00:48:05,960
Speaker 2: the long term. But you know, unfortunately this ends up

816
00:48:06,040 --> 00:48:09,640
Speaker 2: rolling out into so many things about human experience and

817
00:48:09,719 --> 00:48:12,400
Speaker 2: human perception. Is that we are so focused on the

818
00:48:12,400 --> 00:48:14,800
Speaker 2: short term and we don't think about the long term.

819
00:48:15,040 --> 00:48:16,840
Speaker 3: So we've just been talking about a lot of reasons

820
00:48:16,880 --> 00:48:20,000
Speaker 3: that it might actually be bad and not as clever

821
00:48:20,080 --> 00:48:23,480
Speaker 3: as it first seems to try to leverage knowledge of

822
00:48:23,520 --> 00:48:28,800
Speaker 3: ambiguity aversion offensively and negotiations. But I think one way

823
00:48:28,880 --> 00:48:32,240
Speaker 3: that it can definitely be useful is to have knowledge

824
00:48:32,239 --> 00:48:35,840
Speaker 3: of ambiguity aversion to use defensively to think about in

825
00:48:35,960 --> 00:48:41,800
Speaker 3: analyzing your own behavior and being aware of your bias

826
00:48:41,840 --> 00:48:45,440
Speaker 3: to avoid ambiguity and sort of check yourself and think, like,

827
00:48:45,520 --> 00:48:48,520
Speaker 3: wait a minute, am I actually making the right decision here?

828
00:48:49,040 --> 00:48:52,080
Speaker 3: Is this actually what's rational? Or am I just having

829
00:48:52,160 --> 00:48:56,880
Speaker 3: an irrational bias against situations with unknown probabilities that are

830
00:48:57,800 --> 00:49:00,840
Speaker 3: having an outsized scariness in my mind because of the

831
00:49:00,880 --> 00:49:04,520
Speaker 3: ambiguity involved. Like the example we talked about earlier with

832
00:49:04,560 --> 00:49:07,319
Speaker 3: people being afraid to try new experiences because there's some

833
00:49:07,360 --> 00:49:08,400
Speaker 3: amount of ambiguity.

834
00:49:08,800 --> 00:49:11,600
Speaker 2: Yeah, absolutely, there Again, there's always ambiguity and any new

835
00:49:11,640 --> 00:49:14,160
Speaker 2: experience there there's so many ways that could go wrong,

836
00:49:14,200 --> 00:49:15,759
Speaker 2: but there are so many ways that could go right.

837
00:49:16,600 --> 00:49:18,640
Speaker 2: You know, a lot of us do tend to focus

838
00:49:18,680 --> 00:49:19,960
Speaker 2: more on the ways that could go wrong.

839
00:49:20,200 --> 00:49:22,520
Speaker 3: The devil, you know, is not necessarily better.

840
00:49:24,239 --> 00:49:26,120
Speaker 2: Yeah, the devil you don't know, It could be really fun,

841
00:49:26,600 --> 00:49:27,320
Speaker 2: give them a shot.

842
00:49:27,440 --> 00:49:30,960
Speaker 3: Sometimes the devil you know could be like a chaotic

843
00:49:31,000 --> 00:49:33,920
Speaker 3: good devil. Those those exist, right are there?

844
00:49:33,960 --> 00:49:38,359
Speaker 2: So you know, maybe under the new it's possible under

845
00:49:38,400 --> 00:49:40,800
Speaker 2: the new rules, but it's.

846
00:49:40,640 --> 00:49:42,359
Speaker 3: Actually just a friendly tee fling that.

847
00:49:42,280 --> 00:49:44,840
Speaker 2: You can get mistigo. You could have a kat a

848
00:49:44,840 --> 00:49:47,960
Speaker 2: good teethling. There you go, it's entirely possible. All right.

849
00:49:48,000 --> 00:49:50,040
Speaker 3: Well, that's going to do it for part one of

850
00:49:50,239 --> 00:49:53,560
Speaker 3: our look at ambiguity aversion, But we're going to be

851
00:49:53,600 --> 00:49:56,640
Speaker 3: back again next time to talk about how this applies

852
00:49:56,680 --> 00:50:00,440
Speaker 3: to some other domains of life and maybe some subsequent research,

853
00:50:00,600 --> 00:50:03,839
Speaker 3: I might get some into the different ways that the

854
00:50:03,880 --> 00:50:09,120
Speaker 3: ambiguity version observations have been interpreted in terms of different

855
00:50:09,120 --> 00:50:11,880
Speaker 3: types of decision theory, like is it an error or

856
00:50:11,920 --> 00:50:14,680
Speaker 3: is it actually a type of rationality that needs to

857
00:50:14,719 --> 00:50:17,560
Speaker 3: be better understood? So yeah, well we'll talk about that

858
00:50:17,600 --> 00:50:18,520
Speaker 3: kind of stuff next time.

859
00:50:18,760 --> 00:50:22,240
Speaker 2: Who knows what angles we'll actually discuss in the next episode.

860
00:50:22,280 --> 00:50:24,640
Speaker 2: You can tune in to find out in the meantime,

861
00:50:25,000 --> 00:50:27,160
Speaker 2: certainly right in we'd love to hear from you, and

862
00:50:27,239 --> 00:50:29,360
Speaker 2: we'd like to remind you that Stuff to Blow Your

863
00:50:29,400 --> 00:50:31,799
Speaker 2: Mind is primarily a science and culture podcast, with core

864
00:50:31,840 --> 00:50:35,640
Speaker 2: episodes on Tuesdays and Thursdays, short form episode on Wednesdays

865
00:50:35,640 --> 00:50:37,719
Speaker 2: and on Fridays. We set aside most serious concerns to

866
00:50:37,760 --> 00:50:40,719
Speaker 2: just talk about a weird film on Weird House Cinema.

867
00:50:40,960 --> 00:50:44,440
Speaker 3: Huge thanks as always to our excellent audio producer JJ Posway.

868
00:50:44,760 --> 00:50:46,400
Speaker 3: If you would like to get in touch with us

869
00:50:46,400 --> 00:50:48,840
Speaker 3: with feedback on this episode or any other, to suggest

870
00:50:48,840 --> 00:50:50,680
Speaker 3: a topic for the future, or just to say hello,

871
00:50:50,840 --> 00:50:53,520
Speaker 3: you can email us at contact at stuff to Blow

872
00:50:53,520 --> 00:51:01,800
Speaker 3: your Mind dot com.

873
00:51:01,960 --> 00:51:04,920
Speaker 1: Stuff to Blow Your Mind is production of iHeartRadio. For

874
00:51:05,000 --> 00:51:07,759
Speaker 1: more podcasts from my heart Radio, visit the iHeartRadio app,

875
00:51:07,920 --> 00:51:25,360
Speaker 1: Apple podcasts, or wherever you listen to your favorite shows.