When Prime Minister Boris Johnson entered 10 Downing Street in July 2019, he was required to write the “letters of last resort” – his instructions should the UK be hit by a nuclear strike. At the same time, the 1987 INF Treaty, which banned an entire category of weapons, is now officially over, and the prospects of a failure of the 2020 NPT Review Conference and the non-extension of the 2010 New START Treaty next year saturate nuclear discussions.
As legitimately preoccupying as these ongoing events are, the exclusive focus on them obscures what happened 75 years ago, perpetuating an asymmetrical memory of the atomic bombings of World War II, privileging Hiroshima. Let’s not forget that on August 9, 1945, a 21-kiloton atomic bomb levelled the Japanese city of Nagasaki. It was the third atomic explosion in the history of humankind, with more than 2,000 others to come.
Nuclear weapons and citizens’ knowledge
In a context in which every nuclear-weapon state is engaged in large and long-term investments to perpetuate its nuclear arsenal for more than half a century, and citizens outside the UK have not explicitly been given choices on those policies which will impact them for generations, it is crucial to know what citizens know about nuclear-weapons policy.
This is all the more important as their consent is assumed in at least three ways:
Given that we still have no protection against a nuclear attack, be it deliberate or accidental, they are assumed to accept to be potential victims of nuclear harm, coming from a nuclear-armed adversary or from an accident in the nuclear arsenal in the country where they live.
If they are residents of nuclear weapon states, and possibly host states, they are also expected to fund nuclear weapons as taxpayers.
If they are citizens of nuclear weapons possessing states, their voice is implicitly mobilised in support of any nuclear strike the leader may decide to initiate.
For citizens’ consent to be meaningful as justification for a policy, it has to be informed. Is it?
We are answering this question based on an unprecedented large-scale survey of citizens’ knowledge and attitudes in nine countries of the EU and NATO: the two nuclear-weapon states (France and the UK), the five states hosting US nuclear weapons on their territory (Belgium, Germany, Italy, the Netherlands and Turkey) and two countries that have been vocal on nuclear weapons policy or ballistic missile defence (Sweden and Poland).
The survey took place in June 2018 and is based on a representative panel of 7,000 citizens aged 18 to 50. We asked basic questions about which countries possess nuclear weapons, how many of them there are in the world and in the respondent’s country, the effects of a nuclear-weapon explosion, and how many nuclear weapons tests have been conducted since 1945.
Hiroshima and Nagasaki
The first striking set of findings has to do with citizens’ knowledge regarding nuclear weapons, which is more limited than expected. For instance, less than two thirds of respondents knew that Hiroshima and Nagasaki were the two cities hit by atomic weapons in World War II – the numbers vary from 45.7% in Belgium to 74.4% in Italy, with results below 50% in France and the Netherlands. The atomic bombings of World War II are important reference points in debates about the historical legitimacy of today’s nuclear weapons policies and in all the surveyed countries, Nagasaki is systematically less cited than Hiroshima by a significant margin.
The only good news here is that the likelihood of a correct answer increases with age, which creates a hope that respondents may learn about this later in life. However, only 25% of respondents give the adequate order of magnitude of casualties for Hiroshima and Nagasaki (150,000 to 250,000; ranging from 19.6% in France to 30.6% in Sweden) with 28% saying they don’t know, 21% significantly overestimating and 26% significantly underestimating it.
For sure, Hiroshima was the first of the two bombings and immediate casualties and land destroyed are higher. However, remembering Hiroshima as the symbol of atomic bombings and neglecting Nagasaki has fundamental implications on how one sees the meaning of and possibilities within the nuclear age for at least three reasons. First, Hiroshima is still associated with the U.S. official narrative born in 1947: those bombings resulted from planning for a lesser evil in order to save American lives that would have been lost in a ground invasion of Japan, should the war have continued. The fabrication and falsehood of this narrative have been documented by careful historical research.
The need to justify those bombings was felt the day before Nagasaki and it was then that the official rationale for the two bombings started to be crafted. It is only after Nagasaki that President Truman issued his first affirmative command: no more strikes without his explicit consent. Second, while the raid to Hiroshima has been shown to be well planned and executed, the opposite is true for the one to Nagasaki, which was originally just a secondary target and ended up being struck by a bomber heading for Kokura.
The causes of that in flight change remain debated but, even years later, the head of the Manhattan project, General Groves, was not able to understand why Nagasaki ever became the target. Even after target change, ground zero ended up being some three-quarters of a mile off target. Third, to paraphrase what Telford Taylor, the chief US prosecutor at the Nuremberg trials, wrote in 1970: “The rights and wrongs of Hiroshima are debatable,” and are still debated by serious scholars, “but I have never heard a plausible justification of Nagasaki”. Whether or not Nagasaki features in the narrative of the atomic bombings of World War II shifts the historical and military discussion of the bombings from a discussion of strategic rationality, calculated decision-making and military planning and implementation to a discussion of errors, contingency, bad luck for Nagasaki and good luck for Kokura. It also considerably modifies the discussion about whether those bombings were justified.
A similar lack of knowledge is visible about the history of nuclear testing and the current situation regarding nuclear weapons.
When asked how many nuclear weapon tests have been conducted since 1945, 65% of respondents say they don’t know (from 52.5% in Poland to 78.4% in France, which means more than half in every country surveyed) and less than 2% guess the right number. Interestingly, 27.9% significantly underestimate the number of tests conducted so far, offering numbers between 0 and 1,000 – i.e. less than half of the actual number of tests – while only 2.8% massively overstate, giving an answer at least one order of magnitude too high.
Less than 22% of respondents overall (ranging from 19% in Turkey to 27% in Sweden) identify the right order of magnitude when it comes to the number of nuclear weapons in the world today. The same lack of knowledge is visible when the question is about the number of nuclear weapons in the respondents’ country. On average, 65.6% say they do not know and, if one excludes Poland and Sweden in which roughly half of the respondents do say that there are no weapons on their soil, the rate of respondents who claim not to know reaches 70.4%. Even in the UK and France, where the numbers of weapons are publicly available, only 2.9% and 1.7% of respondents approach the official figures.
Only 3.6% of respondents pick all nine nuclear weapon states (the US, Russia, the UK, France, China, Israel, India, Pakistan, North Korea) without mistakenly adding any non-nuclear state to the list. If one takes into account that the US and Russia account for 92% of the arsenal on the planet, one has to note that only 69% of respondents overall answer both (this goes up to 75% in the UK) and 44.5% of respondents wrongly list Iran as already possessing nuclear weapons (this ranges from 36.7% in France to more than 50% in Turkey and the Netherlands).
Similarly, knowledge of the effects of nuclear weapons explosions are very approximative. For instance, only 57.5% of respondents know that they create mass fires in spite of scholarship establishing that for many years. This is by far best known in the UK where 71.5% of respondents know this as opposed to 53.7% in France and 43.7% in Italy. At the same time, 21% and 33% of respondents tick “hurricanes” and “erosion”, which are not adequate answers.
A post-Cold War effect?
The common assumption of universal loss of knowledge on the part of the post–Cold War, post-testing generation is not confirmed. If one compares the cohort of citizens who were teenagers at the end of the Cold War and the generation that comes after – i.e. the 43 to 50 years old and the 18 to 42 years old in 2018 – it is true that the older cohort is more likely to identify the US and Russia as nuclear weapons states (71.5% of respondents against 61.5% in France; 80 against 73% in the UK; 75 against 67% across countries) and has a better sense that nuclear weapons explosions cause radiation (82.5 against 75.5% in France; 91.5 versus 86.5% in the UK; 89 against 83% overall).
However, there is no significant difference in the awareness of the number of victims of Hiroshima and Nagasaki, of the number of nuclear weapons in the respondent’s country or of the number of nuclear tests since 1945. Unexpectedly limited knowledge is visible across cohorts. 72% of respondents in the 43- to 50-year-old cohort stated that they do not know the number of nuclear tests that occurred in the past compared to 65.6% of the younger group (84% versus 76% in France; no differences between cohorts when the aggregate data (the nine countries combined) are considered). These figures should not be taken, however, as an indicator of the younger cohort as more knowledgeable on the occurrence of nuclear tests: among the respondents who gave a figure, more than 82% in both cohorts underestimated the number of nuclear tests ever carried and offered numbers between zero and one thousand.
It is interesting to note that, when asked about the number of nuclear weapons on the planet, the older generation is more likely to overestimate it and answer “50,000” or “100,000 or more” (5% more overall, 4.5% more in France, 10% more in the UK) which may suggest assumptions of legacy of the Cold War. In the UK, the younger cohort gets the right answer by a margin of almost 6% (23.8% vs 18.1%). Finally, the mistake which consists in mentioning Iran as a nuclear weapon state is less frequent among the post–Cold War generation (in France 34.5 versus 42%, in the UK and overall 43 versus 48%).
Those findings give rise to several reactions.
First, invoking or assuming the informed consent of citizens about nuclear weapons policy seems to be a massive overstatement given our respondents’ level of knowledge.
Second, as a prelude to better nuclear education and clearer choices for our citizens and elected officials, more research is needed on the causes and sources of citizens’ nuclear knowledge. The findings of this survey suggest avenues for further inquiry. Respondents from the UK and Sweden stand out as overall better informed than their counterparts in other countries even though an astonishingly high number of 65.7% of Swedish respondents say that no country has ever given up a nuclear weapon program even though their own country actually has. To what extent did the structures of nuclear knowledge production, legitimation and dissemination in the UK and Sweden contribute to this improvement of public knowledge? What was the impact of the Trident debate? How can we assess the possibility of a post-Cold War or post-testing generational change in knowledge and attitudes beyond the simple division in cohorts we proposed here?
Finally, we should resist the temptation to oppose citizens’ uninformed opinions and experts’ knowledge and feel relieved that the former are not deciding anything on the matter, precisely because of their ignorance.
On the one hand, such a view would be missing the well-documented flaws in expert knowledge, their frequent overconfidence, limited ability to update their established judgements in the face of contradicting new evidence and lack of accountability.
On the other hand, one cannot expect well-informed understanding about nuclear affairs from citizens and blame them for their lack of it. A priori, it seems that citizens’ factual knowledge about nuclear weapons politics and history often reflects shared biases among experts and media pundits at the present time. They are particularly visible in the striking lack of knowledge about past nuclear testing, which is absent from the public conversation, the erroneous and frequent mention of Iran as a possessor of nuclear weapons given the tendency to cover it as “imminent proliferation” since 2006, and the relative ignorance of Nagasaki as opposed to Hiroshima, which became a symbol for the nuclear bombings of World War II.