1
00:00:00,060 --> 00:00:07,620
And I want to introduce Jamie Kwong, who's the Marshall scholar and war studies at King's College, London, and amongst other things,

2
00:00:07,620 --> 00:00:16,380
she'll be talking about information and misinformation, which is a topic that also in another context has been examining as well,

3
00:00:16,380 --> 00:00:23,130
because it can be quite harrowing as we start to think through the power of not just lack of information,

4
00:00:23,130 --> 00:00:31,350
but also the power of misinformation and how that shapes not just the dialogue, but actually has potentially very dangerous consequences.

5
00:00:31,350 --> 00:00:34,680
So welcome, Jamie, and thank you. Thank you so much.

6
00:00:34,680 --> 00:00:42,420
And it's actually great to end on Jeremy's point, just touching on some of these public attitudes, because my talk will focus on that.

7
00:00:42,420 --> 00:00:49,410
But before I begin, I just wanted to thank the organisers for having me, as well as all of my fellow speakers for their fascinating insights.

8
00:00:49,410 --> 00:00:54,540
I hope I can add some additional insights to that. So just a little bit about me.

9
00:00:54,540 --> 00:01:03,780
I'm a research assistant and PhD candidate at King's College, London, where my PhD focuses on US public opinion of North Korean nuclear proliferation.

10
00:01:03,780 --> 00:01:09,480
So I hope to zoom out a bit and talk about public opinion of nuclear weapons issues more broadly here.

11
00:01:09,480 --> 00:01:13,980
So I'm going to spend a lot of time covering three key areas.

12
00:01:13,980 --> 00:01:19,290
So first, thinking about interpreting public opinion polls, what can polls tell us?

13
00:01:19,290 --> 00:01:28,080
And importantly, what can't they tell us? Second, looking at the state of play, so what do we know about public opinion of nuclear issues?

14
00:01:28,080 --> 00:01:34,810
And then to Isabel's point, I thirdly want to talk about kind of contextualising these opinions.

15
00:01:34,810 --> 00:01:38,910
So talking about these non public actors, as I'll be calling them.

16
00:01:38,910 --> 00:01:46,860
So the role of NGOs, advocacy organisations, media policymakers, the like in reporting and shaping public opinion.

17
00:01:46,860 --> 00:01:53,370
So my bottom line up front here is that the public holds complex views about nuclear weapons issues.

18
00:01:53,370 --> 00:01:59,700
And so therefore we as experts and scholars need to be very thoughtful about how we analyse public

19
00:01:59,700 --> 00:02:06,690
opinion polls and should be wary about how various actors influence and report public opinion.

20
00:02:06,690 --> 00:02:12,090
So for us to begin with interpreting public opinion polls, what can and can't they tell us?

21
00:02:12,090 --> 00:02:17,340
So there's three things I want to consider here. So first, thinking about question types.

22
00:02:17,340 --> 00:02:22,080
So ultimately there's different attitudes that we can measure through public opinion polling.

23
00:02:22,080 --> 00:02:24,110
So first we have knowledge questions.

24
00:02:24,110 --> 00:02:31,530
So this again gets back to Jeremy's presentation, asking the public objective questions with a right and a wrong answer.

25
00:02:31,530 --> 00:02:36,780
So something that comes up in my research, for example, is what country does Kim Jong Il lead?

26
00:02:36,780 --> 00:02:40,230
And so these questions, unfortunately, are few and far between.

27
00:02:40,230 --> 00:02:48,510
But they're really helpful when they do pop up to gauge the public's actual understanding of the issues they're asked to express opinions about.

28
00:02:48,510 --> 00:02:50,980
And second, we have awareness questions.

29
00:02:50,980 --> 00:02:58,680
So this kind of helps gauge public salience of the topics, how much the public is paying attention to a particular issue,

30
00:02:58,680 --> 00:03:03,000
which again, hopefully builds on Jeremy's point of where are they getting this information?

31
00:03:03,000 --> 00:03:09,570
How much does it relate to media coverage? And third, you can also think about support for different policies.

32
00:03:09,570 --> 00:03:14,490
And so this often in the literature comes up as support for the use of force.

33
00:03:14,490 --> 00:03:21,750
But there's other formulations. So, for example, do you think your country should join the treaty on the prohibition of nuclear weapons or the TPN,

34
00:03:21,750 --> 00:03:25,610
as Nick Ritchie addressed in his his talk earlier?

35
00:03:25,610 --> 00:03:33,660
And so I think those questions, these kind of attitudes questions are the most commonly thought of metrics when you talk about public opinion polling.

36
00:03:33,660 --> 00:03:40,890
Right. It's called public opinion polling for a reason, but it's important to remember that there are different metrics that we can gauge.

37
00:03:40,890 --> 00:03:46,420
And so it's critical not to overstate what each specific question is measuring.

38
00:03:46,420 --> 00:03:53,200
So second type of consideration to think about when looking at public opinion polls is longitudinal data.

39
00:03:53,200 --> 00:04:00,780
So I'm just going to quickly pull up a slide here. Oops.

40
00:04:00,780 --> 00:04:05,010
There we are. Hoping that's popping up now,

41
00:04:05,010 --> 00:04:10,500
but it's so it's really helpful for analysis purposes when polling companies ask the same question

42
00:04:10,500 --> 00:04:16,590
repeatedly because it allows for a fairly robust assessment of changes in public opinion over time.

43
00:04:16,590 --> 00:04:25,050
And having this trend data helps us pinpoint fluctuations and allow us to explore why there might be particular spikes or dips,

44
00:04:25,050 --> 00:04:27,180
in particular attitude metrics.

45
00:04:27,180 --> 00:04:34,890
So, for example, again, this is one that comes up in my research, but you'll see that this looks at public threat perception of North Korea over time.

46
00:04:34,890 --> 00:04:42,000
And that box and red highlights a particular spike when the public found North Korea to be a very serious threat.

47
00:04:42,000 --> 00:04:47,880
And if you're aware of the North Korean nuclear timeline, that twenty seven year in July,

48
00:04:47,880 --> 00:04:54,640
North Korea tested an intercontinental ballistic missile and in September, it tested a North Korean nuclear weapon.

49
00:04:54,640 --> 00:05:02,550
And so you do see kind of a correlation there. So that's something that we can help analyse to see these trends over time.

50
00:05:02,550 --> 00:05:08,100
And ultimately, this longitudinal data allows us to draw more reliable conclusions about public

51
00:05:08,100 --> 00:05:12,810
attitudes and give this more full and comprehensive picture of public opinion.

52
00:05:12,810 --> 00:05:17,190
And so it's a much stronger and better practise than relying on one discrete public

53
00:05:17,190 --> 00:05:22,320
opinion poll and asserting that that poll accurately reflects public attitudes.

54
00:05:22,320 --> 00:05:27,720
Which leads me to my final point and perhaps most importantly, question wording matters.

55
00:05:27,720 --> 00:05:31,770
So actually, considering what the words in the question say.

56
00:05:31,770 --> 00:05:40,890
So, for example, here we're just seeing do you think North Korea presents a serious or non serious threat to the US?

57
00:05:40,890 --> 00:05:47,550
That's distinct from saying, do you think North Korea's nuclear weapons programme poses a serious threat to US interests?

58
00:05:47,550 --> 00:05:55,440
And so you generally find that the one when the word nuclear is mentioned, threat perception increases, which makes intuitive sense.

59
00:05:55,440 --> 00:06:02,220
But the point being phrasing can prime respondents to answer specific and often biased ways.

60
00:06:02,220 --> 00:06:08,130
And second, to think about here as well are what assumptions are we making when we're asking these questions?

61
00:06:08,130 --> 00:06:14,580
So it's critical, again, to Jeremy's point that we don't assume that the public has a fundamental knowledge or

62
00:06:14,580 --> 00:06:21,270
understanding of the context and even just the factual notions around nuclear weapons issues.

63
00:06:21,270 --> 00:06:26,640
And so, for example, asking the public if they support the Tea Party and W is less helpful than providing

64
00:06:26,640 --> 00:06:31,560
a description of the TPN W and then asking whether or not the public support it.

65
00:06:31,560 --> 00:06:41,310
However, it's critical to provide objective context to truly inform the public and gauge their nonbiased responses to the information presented,

66
00:06:41,310 --> 00:06:49,020
which again gets back to the phrasing and potential priming effects. So the large takeaway here is that when we're analysing public opinion polls,

67
00:06:49,020 --> 00:06:55,710
we should be very wary of any polls that don't provide objective context or don't provide any context at all.

68
00:06:55,710 --> 00:07:02,070
And it's especially problematic if we rely on single data points or polls that are continually making these

69
00:07:02,070 --> 00:07:08,890
assumptions without either measuring public knowledge or providing this additional contextual information.

70
00:07:08,890 --> 00:07:13,320
OK, so to get off my scholarly, how do we think about these polls?

71
00:07:13,320 --> 00:07:17,460
What do we actually know about public opinion of nuclear issues?

72
00:07:17,460 --> 00:07:21,840
So importantly, it's first important to note that public,

73
00:07:21,840 --> 00:07:27,240
the public rarely considers nuclear issues as the most important problem facing their country.

74
00:07:27,240 --> 00:07:31,140
They're much more likely to give attention to domestic issues.

75
00:07:31,140 --> 00:07:36,150
So you might interpret this as nuclear weapons are unlikely to be a central voting issue.

76
00:07:36,150 --> 00:07:41,940
But nevertheless, we care about public opinion and understanding of nuclear issues for many of the reasons discussed.

77
00:07:41,940 --> 00:07:44,910
I'm happy to go into them further in the Q&A,

78
00:07:44,910 --> 00:07:52,770
but what exactly do we know about these opinions so we can divide public opinion of these issues into a few big buckets?

79
00:07:52,770 --> 00:08:00,390
And so these aren't all encompassing, but it kind of gives a sense of the key areas where contemporary polling tends to focus.

80
00:08:00,390 --> 00:08:04,260
And no surprise, I'll look at three of them. So first, nuclear use.

81
00:08:04,260 --> 00:08:11,160
Second, nuclear capabilities and third, disarmament slash the TPN w so to the first,

82
00:08:11,160 --> 00:08:14,310
a lot of the experimental survey work that occurs in the literature.

83
00:08:14,310 --> 00:08:23,010
And by this I mean researchers conducting their own surveys to test for very clear causal relationships between public attitudes and specific factors.

84
00:08:23,010 --> 00:08:28,290
A lot of these surveys are looking at the public attitudes around the use of nuclear weapons.

85
00:08:28,290 --> 00:08:37,680
And so key scholars here, Scott Sagan and Ben Valentino, who have conducted a series of studies showing that despite the tradition of nuclear non use,

86
00:08:37,680 --> 00:08:45,870
the US public in particular is willing to support the use of nuclear weapons to preserve American lives and win a war quickly.

87
00:08:45,870 --> 00:08:51,300
And this attitude is tempered less by the nuclear taboo.

88
00:08:51,300 --> 00:08:57,780
Another key point in the nuclear literature and more so by fear of retribution and setting a negative precedent.

89
00:08:57,780 --> 00:09:02,800
Additional work by some scholars named Cocain Wells have gone further to show that this attitude.

90
00:09:02,800 --> 00:09:06,910
Is also significantly tempered when given vivid information about the physical

91
00:09:06,910 --> 00:09:11,110
effects of a nuclear attack and the possible nuclear risks and retaliation,

92
00:09:11,110 --> 00:09:17,080
so again, this point of educating the public helps them make more informed decisions and attitudes.

93
00:09:17,080 --> 00:09:20,620
Second, thinking about public attitudes around nuclear capabilities.

94
00:09:20,620 --> 00:09:27,880
So this is probably the biggest bucket and can take many forms, but often it relates to particularly policy, relevant topics.

95
00:09:27,880 --> 00:09:37,570
So, for example, US public attitudes on the replacement ICBM programme, which is called the ground based strategic deterrent or for short.

96
00:09:37,570 --> 00:09:42,250
So this is an increasingly contested topic amongst the expert community and within

97
00:09:42,250 --> 00:09:48,190
Congress as to whether or not the United States should continue to pursue this programme,

98
00:09:48,190 --> 00:09:53,890
given its costs, potential incomplete consideration of other alternatives, like I won't get into that there.

99
00:09:53,890 --> 00:09:58,990
But the Federation of American Scientists has told the public what they think about this,

100
00:09:58,990 --> 00:10:08,870
and they found that most support alternatives to leave and go so far as to support phasing out ICBMs altogether.

101
00:10:08,870 --> 00:10:13,750
So there's not much polling yet for comparison to this poll,

102
00:10:13,750 --> 00:10:21,010
but it suggests that this can be an area for future research and also to think about capabilities from a non US perspective,

103
00:10:21,010 --> 00:10:25,930
we can look at South Korean public attitudes on a national nuclear arsenal.

104
00:10:25,930 --> 00:10:33,520
And so we tend to see relatively high levels of support amongst the South Korean public for developing its own nuclear arsenal.

105
00:10:33,520 --> 00:10:36,430
But again, once you provide additional context,

106
00:10:36,430 --> 00:10:42,220
that support is significantly tempered when questions include discussion of the economic and security implications

107
00:10:42,220 --> 00:10:51,190
of doing so and finally thinking about public attitudes around disarmament in general and the TPN in particular,

108
00:10:51,190 --> 00:10:57,970
I want to kind of address this in a few different ways, because I know it's a topic that has come up already this morning.

109
00:10:57,970 --> 00:11:02,890
So one way to think about this is through the polling work of advocacy organisations.

110
00:11:02,890 --> 00:11:09,640
So, for example, the international campaign to abolish nuclear weapons, I can does some original polling on this topic.

111
00:11:09,640 --> 00:11:15,220
And so their polls show that majorities in NATO states support joining the TPN W

112
00:11:15,220 --> 00:11:21,280
and most support even doing so unilaterally and in spite of US pressure not to,

113
00:11:21,280 --> 00:11:25,120
then we can think about experimental survey work on the TPN.

114
00:11:25,120 --> 00:11:29,320
So a lot of this work is done by Rebecca Gibbons' and Stephen Hertzog.

115
00:11:29,320 --> 00:11:37,660
And they show that in the US elite cues, so arguments by people in the government against the treaty and social cues saying,

116
00:11:37,660 --> 00:11:45,400
you know, my peer also is against the treaty because the US public support for the TPN to decrease.

117
00:11:45,400 --> 00:11:47,650
However, interestingly, in Japan,

118
00:11:47,650 --> 00:11:56,530
their work shows that public support for the TPN W remains fairly high and constant when introduced to arguments by the government against the treaty,

119
00:11:56,530 --> 00:12:01,270
as I'm pretty sure Professor Northmoor will touch on in his remarks.

120
00:12:01,270 --> 00:12:08,920
And a final way to think about public attitudes on disarmament is analysis of existing public opinion polling data.

121
00:12:08,920 --> 00:12:12,010
So one of my colleagues at King's College London, Amelia Morgan,

122
00:12:12,010 --> 00:12:18,400
has done this in a recent piece where she shows that the public supports public support

123
00:12:18,400 --> 00:12:24,610
for the abstract concept of disarmament is much higher than support for the TPN itself.

124
00:12:24,610 --> 00:12:33,160
So kind of thinking about these different ways of how we can interpret public opinion of disarmament reminds us that to the points outlined above,

125
00:12:33,160 --> 00:12:35,950
it's important not to take polls at their face value.

126
00:12:35,950 --> 00:12:45,130
And in particular, you need to consider these non public actors involved in shaping, collecting information about and reporting public opinion.

127
00:12:45,130 --> 00:12:52,390
But just before I conclude and discuss those actors quickly, I want to underscore with these findings of these different attitudes buckets,

128
00:12:52,390 --> 00:12:58,960
as I've called them collectively tell us, and that's that public opinion of nuclear weapons issues is complex.

129
00:12:58,960 --> 00:13:05,020
So it's something I discuss often with my colleague Amelia is something that her piece really excellently demonstrates.

130
00:13:05,020 --> 00:13:13,450
But it's unfair for us as nuclear weapons experts to expect the public to have more clear and concise views of nuclear issues than we do.

131
00:13:13,450 --> 00:13:16,930
Right. We've spent all morning having contesting perspectives of this.

132
00:13:16,930 --> 00:13:21,290
So why should we expect the public to have figured it out any better than we have?

133
00:13:21,290 --> 00:13:25,100
So finally, just to conclude, I want to touch on these non-public actors,

134
00:13:25,100 --> 00:13:31,850
kind of a cheeky little term I came up with, but those that can kind of influence and shape public opinion.

135
00:13:31,850 --> 00:13:41,360
So I'm thinking any non public actors, so NGOs, advocacy organisations, the media, subject matter experts, policymakers.

136
00:13:41,360 --> 00:13:47,900
Again, let me touch on three considerations here. So a lot of academic work focuses on what's called elite use.

137
00:13:47,900 --> 00:13:56,780
So because foreign policy issues aren't proximate to the public, the public instead relies on cues from elites about how to think about these issues.

138
00:13:56,780 --> 00:14:00,470
So there's exceptions for when this is effective and not effective.

139
00:14:00,470 --> 00:14:05,930
But generally speaking, we tend to see that elite curlicues influence public opinion again.

140
00:14:05,930 --> 00:14:07,760
Makes intuitive sense, right?

141
00:14:07,760 --> 00:14:15,200
If you don't know much about a particular foreign policy issue, you're likely to consider in similar terms to an elite that you respect.

142
00:14:15,200 --> 00:14:26,150
So whether that's a New York Times columnist who you turn to or an elected official, second thinking about the media and a potential feedback loop.

143
00:14:26,150 --> 00:14:32,480
So we actually see that a lot of the regular polling that's done, especially in the US, is by media organisations.

144
00:14:32,480 --> 00:14:41,120
So you'll see on the Roper Centre, for instance, which houses a huge collection of public opinion polls, you'll see ABC and The Washington Post.

145
00:14:41,120 --> 00:14:44,870
You'll see CBS and The New York Times collecting these polls.

146
00:14:44,870 --> 00:14:52,040
So in doing so, there's this inherent feedback loop where questions reflect particularly salient news topics.

147
00:14:52,040 --> 00:14:56,990
And it's the same case for your classic polling organisations like Gallup and Pew.

148
00:14:56,990 --> 00:15:04,430
So we're seeing that these polling organisations are interested in knowing what the public thinks about the hot topics of the day.

149
00:15:04,430 --> 00:15:08,750
And that's partly, I think, why we don't see consistent polling on nuclear weapons issues.

150
00:15:08,750 --> 00:15:13,640
Right. It's as episodic as is the news coverage of these issues, which again,

151
00:15:13,640 --> 00:15:18,050
is problematic as analysts when we want to track these attitudes and better understand

152
00:15:18,050 --> 00:15:24,740
what the public thinks of nuclear issues in non salient times or in normal circumstances.

153
00:15:24,740 --> 00:15:27,110
And then finally, just to conclude,

154
00:15:27,110 --> 00:15:36,020
I think we need to think about these non-public actors interests and how that might be influencing not only public opinion,

155
00:15:36,020 --> 00:15:44,040
but reporting of public opinion in particular. And so a lot of these non public actors are purposely advancing particular perspectives.

156
00:15:44,040 --> 00:15:50,960
So, for example, advocacy organisations, it's the whole point right there advocating a particular position.

157
00:15:50,960 --> 00:15:55,950
And so in promoting interests, this is typically where we see.

158
00:15:55,950 --> 00:16:02,790
The most priming from question phrasing, so a question is skewed towards a particular position and so that, for example,

159
00:16:02,790 --> 00:16:06,900
could mean that a question doesn't provide an objective account of a debate on

160
00:16:06,900 --> 00:16:11,610
a particular issue and instead leads respondents in a particular direction.

161
00:16:11,610 --> 00:16:18,000
So that way, the organisation conducting the poll kind of gets the the results that they want.

162
00:16:18,000 --> 00:16:25,920
And so this often results in biased reporting and it becomes a particular issue when these groups are particularly skilled at public messaging.

163
00:16:25,920 --> 00:16:32,550
So that reporting often gets misrepresented as a comprehensive review of public opinion on a particular issue.

164
00:16:32,550 --> 00:16:41,790
But I'd argue that injecting non-public actor's interests into public polling is a disservice not only to the public, but also to policymakers.

165
00:16:41,790 --> 00:16:47,880
We should be striving to objectively consider what the public thinks about nuclear issues and try to better

166
00:16:47,880 --> 00:16:54,450
understand the complexity of those issues to ensure that our nuclear policies are as representative as possible.

167
00:16:54,450 --> 00:17:00,230
And so I'll just go ahead and conclude there and turn it back over to Isabella. Thanks so much again.

168
00:17:00,230 --> 00:17:06,590
Jamie, thank you so much for that and also how you helped us think through the nature of polling and its limitations and then,

169
00:17:06,590 --> 00:17:13,040
of course, the content or what is revealed through polling and then the cautions about even if we have those poll results,

170
00:17:13,040 --> 00:17:17,570
we have to also be very careful and do our due diligence in terms of who's doing the

171
00:17:17,570 --> 00:17:22,100
polling and what other kinds of agendas might be influencing what is then often

172
00:17:22,100 --> 00:17:28,160
perceived as a empirical result or a more powerful result than is actually the case

173
00:17:28,160 --> 00:17:33,257
if you scratch below the surface and went through the methodology that you did. So thank you so much for that.