Everyone working in the field of nuclear weapons claims to be reducing strategic risk as a basic objective, but people often end up working against one another’s priorities, even when they consider themselves to be on a similar ‘side’. There is significant scope for greater appreciation of the merits and risks of differing approaches to nuclear risk reduction, focusing on the available evidence and its implications, the coherence, plausibility and robustness of the models of change different actors seem to follow, and the level of cooperation between the groups involved. We need to be cognisant of the many challenges to collaboration, but the circumstances in which we are currently operating point to the significant benefits from efforts to improve it.
In this post, we[1] present an idea for bringing together different groups, aiming to improve the political impact of the global nuclear policy community as a whole by improving the collective focus on strategic impact. We are currently exploring opportunities for collaboration within the established nuclear policy field, improving access for people from other disciplines, and working to further refine a potential plan for this project. We’re posting ideas on these matters to the EA forum in the hope of receiving critical and constructive feedback; comments on the general idea as well as on any of the more concrete points are extremely welcome!
What? - Project outline
Our proposal is to draw nuclear weapons policy groups (NWPGs) within nuclear armed states and their alliances into collective processes to explore better coordination and potential collaboration on joint projects. We focus on groups which look to achieve impact within political and official circles to drive effective nuclear risk reduction and encourage multilateral nuclear disarmament. Through an interplay of stakeholder engagement processes and rigorous background research, we would synthesise knowledge/experience/data from different NWPGs, help them develop their theories of change, identify and propose triaging of goals for NWPGs, and form hypotheses and methods to test the assumptions behind the theories of change. Insights from these investigations would then serve to help inform existing and new nuclear risk reduction efforts, so as to increase the political impact of the global nuclear policy community as a whole.
Why? - The project rationale
If this project is successful it will not only improve the impact of civil society organisations working for nuclear risk reduction and disarmament, but also lead by example in encouraging governments to better manage their differences and collaborate in tackling multipolar traps (multidimensional coordination problems) associated with global security and broader global catastrophic risk.
Below, we first briefly explain why we believe more coordination between NWPGs is needed in general, and we then list a few features of the current situation that suggest that greater collaboration is particularly worthwhile at this point in time.
The basic case for coordination and synthesis
The nuclear policy community is well connected, at domestic and international level, and populated by well-informed and highly skilled operators. But the divisions between different nuclear weapons policy groups on models of change and focus, a common choice faced being to buy into a particular network’s model or be seen as hostile, appear to discourage effective coordination.
Some of these divisions often feel intractable. For example, it would seem impossible to achieve collaboration between campaign groups focused on binary public messaging that declares nuclear weapons to be immoral and illegal and those looking to manage nuclear risk within the current system of deterrence, with a long term objective to move to other forms of international relations. Different parts of civil society react very differently to the dominant narrative within nuclear armed states that supports the belief that greater credibility of the nuclear threat delivers more stability. As a result, such groups engage in confrontational communication, a situation that simply re-enforces the status quo. Short and long term objectives and tactics are confused, obscuring shared objectives. Opportunities to collaborate to manage the complex polarity between deterrence and disarmament, and to achieve synergies of impact on different parts of the change model are missed.
Another challenge is the competition between groups for funding and impact that tends towards competing models of change to emphasise unique selling points. Inefficient communication can lead to duplication. But we believe that a tendency to attach to specific prior policy positions and a dynamic that involves continual statement and restatement of competing positions can diminish the overall impact of the nuclear weapons policy community on nuclear risk reduction, as well as hindering the effective pursuit of disarmament and improvements to global coordination.
Collective management of the policy tensions involved in arms control and disarmament could improve outcomes. We would like to create fora that bring different groups together without losing their particular identities, helping them coordinate and align their efforts to push jointly towards shared goals whilst respecting differences. This would give nuclear weapons policy groups the opportunity to learn from each other, and create a space to further develop knowledge and enhance strategies for action through discussion and exchange[2]. Through this project, we would also explore how groups could best employ rigorous and effective tools to scrutinise and constructively discuss the different strategies currently pursued in the nuclear risk reduction field. Lastly, the exchanges encouraged by the project might result in joint campaigns or interventions, pooling resources and capabilities from different groups to achieve shared goals.
The urgency of the moment
While the need for effective nuclear risk reduction efforts has been with us ever since the development of the first nuclear bomb in 1945, we would argue that it is particularly great at the present moment. First, the amount of funding going to groups working on nuclear non-proliferation and disarmament has been falling sharply over the last few years, which we believe underlines the need for efficiency and effectiveness in nuclear risk reduction efforts[3]. Second, the current geopolitical climate, ongoing conflicts involving nuclear weapons states (Ukraine, Gaza), and the erosion of the regime to contain the production and deployment of nuclear weapons[4] all contribute to particularly high levels of nuclear risk today (with the Doomsday clock being set at 90 seconds to midnight, the closest it has ever been). And third, the developments just mentioned have raised the salience of nuclear risks amongst the public and in decision-making circles, potentially creating an opening for nuclear risk reduction measures and proposals.
How? - Plan for action
We envisage a process involving the following steps:
Identify and then triage between short term and long term goals, which can include increased awareness of risks, nuclear risk reduction proposals, changes in posture, non-proliferation, disarmament and broader changes in security policies and the international/global security environment.
Develop possible pathways of actions (methods) to achieve these goals, mapping out a number of change trajectories and exploring their qualities and probabilities. Encourage all to engage with an open mind not only with the question of what needs to happen, but also with the type of thinking that is required to implement effective actions and the stories that will support that change of thinking.
Design and critically reflect on methods to measure/assess effectiveness and efficiency of the theories of change (goals & pathways of action) developed before, recognising the challenges and pitfalls in giving confidence to quantitative approaches.
With members across the broader community, rigorously explore approaches to implementing these methods, assess existing and proposed strategies for nuclear risk reduction, producing conclusions that help guide the direction of individual groups within the community.
Continuous feedback loop between analytical investigations (see above) and on-the-ground experience, informing ongoing discussions on effective strategies for the nuclear policy community.
How might this fail? - Challenges and limitations
The initiative described above could fail at several stages of the implementation process:
we might be unable to get buy-in from relevant NWPGs and fail to bring them together in the first place;
the working group sessions with NWPGs might not succeed in generating reliable insights and synergies to enhance effectiveness within and across groups;
insights from the working group sessions might not be taken up; and
we might trigger inadvertent harm that outweighs the benefits from our activities.
Below, we describe the four main challenges we believe we face at all these stages of the process and we provide short explanations of how we intend to respond to them.
Challenges of communication
We expect that it won’t be easy to communicate across the different groups that we’d like to reach with this initiative. Actors in the field care deeply about their goals; this is clearly beneficial and justified in the first place, but it can also have the side-effect of creating entrenched beliefs and defensive attitudes, rendering mutual understanding between groups with different perspectives and strategies for reducing nuclear risks difficult. Based on such entrenched beliefs, some groups may already be quite convinced by the appropriateness of their theory of change; they might thus be unlikely to think that this initiative of jointly reconsidering established strategies holds much value and might show little willingness to engage in open-minded dialogues with us and other groups. This is particularly true if they distrust our proposed methods for investigating potential strategy innovations/improvements, which might lead them to actively oppose our project (based on concerns about adverse effects, as outlined below). Aware of these potential objections and feelings of aversion, we will seek to communicate very clearly that our initiative is not intended as an imposition of a certain perspective or methodology but rather a forum of discussion and a decision aid for NWPGs. We will be open and frank, though, about our attachment to a model of engagement across ideologies and belief systems, alongside clear-eyed awareness of the power struggles and plays within the field of nuclear weapons policy.
A second challenge to effective outreach and communication is the geographic and cultural diversity that we’d ideally like to achieve for this project. We believe that there are large upsides in reaching civil society in as many nuclear weapons states (and allied countries) as possible, because nuclear risk reduction has limited effectiveness if it relies on the actions of only one country. However, we also recognise that there are challenges of logistics and cultural understanding for bringing groups from different countries together in a shared forum and for facilitating effective communication on a topic that is all but trivial and non-important to the participants of the discussion. We hope to leverage our experience in advancing cross-cultural communication and to find partners in some of the other nuclear weapons states in order to deal with these difficulties.
Tractability: challenges of figuring out what actually works
One of the core motivations for this project simultaneously constitutes one of its primary challenges: For all we can tell, figuring out what actually works to reduce nuclear risks effectively is extremely difficult, never mind the challenges of finding some degree of common ground in terms of identifying room for improvement. Moreover, we simply face quite high uncertainty when evaluating the effectiveness of different interventions to impact policy more generally: Goal prioritisation requires that we weigh the importance of different goals, which is mired by ethical and empirical controversies; and the choice of interventions and strategies requires that we attribute policy changes to specific actions, which is difficult in a world marked by non-linear effects, interdependence of causes, and feedback loops[5]. We are devising this initiative in order to make headway on those epistemic and collaborative challenges, and we are driven by the belief that it constitutes a worthwhile attempt: while it certainly is not guaranteed to succeed, it does hold some promise in delivering improvements in the way we approach complexity[6].
Resource constraints
Our initiative is partially motivated by the urgency for effective action created by the shortage of funding, but we also realise that these resource constraints pose limitations and challenges to what a forum for bringing together NWPGs may achieve. If the challenges to communication and tractability can be met, there is still a chance that the insights generated by the forum cannot be put into practice because groups lack the finances to implement new strategies effectively. Although this is an issue, collective strategies that create efficiency savings and more stringently prioritise focused efforts will likely provide more financing per initiative. We also plan to work with nuclear policy funders to integrate their insights and communicate funding requirements and issues across the community.
Funding gaps also raise a further challenge for our efforts to get buy-in from relevant NWPGs: many of them may already be facing a backlog of tasks and projects they would like to implement, raising their reluctance to allocate some of their employees’ time to our initiative. It’s thus imperative that such an initiative has an explicit objective to save time, through shared knowledge acquisition, facilitating more efficient community discussion and avoiding duplication of efforts.
Possible adverse effects
Several of the challenges we described until now don’t just threaten the benefits of our initiative, they also create risks of causing inadvertent harm:
If done or framed in an inconsiderate way, our project might end up increasing competition and hostility between groups (if the initiative is perceived as a forum for ranking and then prioritising resource allocation between different groups and their strategies).
If done naively, rigorous impact evaluations could lead to a narrow-minded and thus inaccurate assessment of different strategies, which could engender unwise action recommendations for allocating funding, designing strategies, and implementing interventions.
If communicated naively, our emphasis on rigorous impact evaluations might reinforce unhelpful norms and behavioural pressures for civil society actors, creating the perceived and/or real need to spend significant resources on “selling” their activities to funders instead of engaging in activities that they believe will actually advance the goal of nuclear risk reduction.[7]
We take these possible negative side-effects extremely seriously and seek to prevent such harm by adopting a risk-aware and self-critical perspective throughout our work on the project. We also encourage participation from those that might harbour considerable scepticism or concern for the approach of our project. For this reason, comments and feedback on this last section would be particularly valuable to us!
“We”, that is: Paul Ingram and Sarah Weiler, in collaboration with others in the nuclear policy field. We have been collaborating for the last year informally to plan and pursue joint projects largely focused on nuclear winter awareness.
Aside from the common-sense case for such discussions, there are numerous studies that indicate that connecting diverse perspectives on tricky questions improves decision quality.
The fall in funding is primarily due to MacArthur’s withdrawal from the field, and is exacerbated by the implosion of FTX and the fall in value of Facebook shares, which reduces Open Philanthropy’s deployable capital.
This includes the breakdown of the START treaty, Russia’s deratification of CTBT, and military modernisation and build-up programmes in several nuclear weapons states.
More extensive discussions of the challenges of uncertainty in policy impact evaluations can be found in Justis 2022, Woods 2022, and Wiblin 2016 (to name just three examples). An early analysis that discusses those challenges specifically with regards to the nuclear policy field is given in Betts 1979.
A similar concern was recently raised in Elizabeth 2023, and is also touched on Karnofsky 2022 (see especially this section). The general problem is well-described by Goodhart’s law (see LessWrong for a quick intro).
More coordinated civil society action on reducing nuclear risk
Everyone working in the field of nuclear weapons claims to be reducing strategic risk as a basic objective, but people often end up working against one another’s priorities, even when they consider themselves to be on a similar ‘side’. There is significant scope for greater appreciation of the merits and risks of differing approaches to nuclear risk reduction, focusing on the available evidence and its implications, the coherence, plausibility and robustness of the models of change different actors seem to follow, and the level of cooperation between the groups involved. We need to be cognisant of the many challenges to collaboration, but the circumstances in which we are currently operating point to the significant benefits from efforts to improve it.
In this post, we[1] present an idea for bringing together different groups, aiming to improve the political impact of the global nuclear policy community as a whole by improving the collective focus on strategic impact. We are currently exploring opportunities for collaboration within the established nuclear policy field, improving access for people from other disciplines, and working to further refine a potential plan for this project. We’re posting ideas on these matters to the EA forum in the hope of receiving critical and constructive feedback; comments on the general idea as well as on any of the more concrete points are extremely welcome!
What? - Project outline
Our proposal is to draw nuclear weapons policy groups (NWPGs) within nuclear armed states and their alliances into collective processes to explore better coordination and potential collaboration on joint projects. We focus on groups which look to achieve impact within political and official circles to drive effective nuclear risk reduction and encourage multilateral nuclear disarmament. Through an interplay of stakeholder engagement processes and rigorous background research, we would synthesise knowledge/experience/data from different NWPGs, help them develop their theories of change, identify and propose triaging of goals for NWPGs, and form hypotheses and methods to test the assumptions behind the theories of change. Insights from these investigations would then serve to help inform existing and new nuclear risk reduction efforts, so as to increase the political impact of the global nuclear policy community as a whole.
Why? - The project rationale
If this project is successful it will not only improve the impact of civil society organisations working for nuclear risk reduction and disarmament, but also lead by example in encouraging governments to better manage their differences and collaborate in tackling multipolar traps (multidimensional coordination problems) associated with global security and broader global catastrophic risk.
Below, we first briefly explain why we believe more coordination between NWPGs is needed in general, and we then list a few features of the current situation that suggest that greater collaboration is particularly worthwhile at this point in time.
The basic case for coordination and synthesis
The nuclear policy community is well connected, at domestic and international level, and populated by well-informed and highly skilled operators. But the divisions between different nuclear weapons policy groups on models of change and focus, a common choice faced being to buy into a particular network’s model or be seen as hostile, appear to discourage effective coordination.
Some of these divisions often feel intractable. For example, it would seem impossible to achieve collaboration between campaign groups focused on binary public messaging that declares nuclear weapons to be immoral and illegal and those looking to manage nuclear risk within the current system of deterrence, with a long term objective to move to other forms of international relations. Different parts of civil society react very differently to the dominant narrative within nuclear armed states that supports the belief that greater credibility of the nuclear threat delivers more stability. As a result, such groups engage in confrontational communication, a situation that simply re-enforces the status quo. Short and long term objectives and tactics are confused, obscuring shared objectives. Opportunities to collaborate to manage the complex polarity between deterrence and disarmament, and to achieve synergies of impact on different parts of the change model are missed.
Another challenge is the competition between groups for funding and impact that tends towards competing models of change to emphasise unique selling points. Inefficient communication can lead to duplication. But we believe that a tendency to attach to specific prior policy positions and a dynamic that involves continual statement and restatement of competing positions can diminish the overall impact of the nuclear weapons policy community on nuclear risk reduction, as well as hindering the effective pursuit of disarmament and improvements to global coordination.
Collective management of the policy tensions involved in arms control and disarmament could improve outcomes. We would like to create fora that bring different groups together without losing their particular identities, helping them coordinate and align their efforts to push jointly towards shared goals whilst respecting differences. This would give nuclear weapons policy groups the opportunity to learn from each other, and create a space to further develop knowledge and enhance strategies for action through discussion and exchange[2]. Through this project, we would also explore how groups could best employ rigorous and effective tools to scrutinise and constructively discuss the different strategies currently pursued in the nuclear risk reduction field. Lastly, the exchanges encouraged by the project might result in joint campaigns or interventions, pooling resources and capabilities from different groups to achieve shared goals.
The urgency of the moment
While the need for effective nuclear risk reduction efforts has been with us ever since the development of the first nuclear bomb in 1945, we would argue that it is particularly great at the present moment. First, the amount of funding going to groups working on nuclear non-proliferation and disarmament has been falling sharply over the last few years, which we believe underlines the need for efficiency and effectiveness in nuclear risk reduction efforts[3]. Second, the current geopolitical climate, ongoing conflicts involving nuclear weapons states (Ukraine, Gaza), and the erosion of the regime to contain the production and deployment of nuclear weapons[4] all contribute to particularly high levels of nuclear risk today (with the Doomsday clock being set at 90 seconds to midnight, the closest it has ever been). And third, the developments just mentioned have raised the salience of nuclear risks amongst the public and in decision-making circles, potentially creating an opening for nuclear risk reduction measures and proposals.
How? - Plan for action
We envisage a process involving the following steps:
Identify and then triage between short term and long term goals, which can include increased awareness of risks, nuclear risk reduction proposals, changes in posture, non-proliferation, disarmament and broader changes in security policies and the international/global security environment.
Develop possible pathways of actions (methods) to achieve these goals, mapping out a number of change trajectories and exploring their qualities and probabilities. Encourage all to engage with an open mind not only with the question of what needs to happen, but also with the type of thinking that is required to implement effective actions and the stories that will support that change of thinking.
Design and critically reflect on methods to measure/assess effectiveness and efficiency of the theories of change (goals & pathways of action) developed before, recognising the challenges and pitfalls in giving confidence to quantitative approaches.
With members across the broader community, rigorously explore approaches to implementing these methods, assess existing and proposed strategies for nuclear risk reduction, producing conclusions that help guide the direction of individual groups within the community.
Continuous feedback loop between analytical investigations (see above) and on-the-ground experience, informing ongoing discussions on effective strategies for the nuclear policy community.
How might this fail? - Challenges and limitations
The initiative described above could fail at several stages of the implementation process:
we might be unable to get buy-in from relevant NWPGs and fail to bring them together in the first place;
the working group sessions with NWPGs might not succeed in generating reliable insights and synergies to enhance effectiveness within and across groups;
insights from the working group sessions might not be taken up; and
we might trigger inadvertent harm that outweighs the benefits from our activities.
Below, we describe the four main challenges we believe we face at all these stages of the process and we provide short explanations of how we intend to respond to them.
Challenges of communication
We expect that it won’t be easy to communicate across the different groups that we’d like to reach with this initiative. Actors in the field care deeply about their goals; this is clearly beneficial and justified in the first place, but it can also have the side-effect of creating entrenched beliefs and defensive attitudes, rendering mutual understanding between groups with different perspectives and strategies for reducing nuclear risks difficult. Based on such entrenched beliefs, some groups may already be quite convinced by the appropriateness of their theory of change; they might thus be unlikely to think that this initiative of jointly reconsidering established strategies holds much value and might show little willingness to engage in open-minded dialogues with us and other groups. This is particularly true if they distrust our proposed methods for investigating potential strategy innovations/improvements, which might lead them to actively oppose our project (based on concerns about adverse effects, as outlined below). Aware of these potential objections and feelings of aversion, we will seek to communicate very clearly that our initiative is not intended as an imposition of a certain perspective or methodology but rather a forum of discussion and a decision aid for NWPGs. We will be open and frank, though, about our attachment to a model of engagement across ideologies and belief systems, alongside clear-eyed awareness of the power struggles and plays within the field of nuclear weapons policy.
A second challenge to effective outreach and communication is the geographic and cultural diversity that we’d ideally like to achieve for this project. We believe that there are large upsides in reaching civil society in as many nuclear weapons states (and allied countries) as possible, because nuclear risk reduction has limited effectiveness if it relies on the actions of only one country. However, we also recognise that there are challenges of logistics and cultural understanding for bringing groups from different countries together in a shared forum and for facilitating effective communication on a topic that is all but trivial and non-important to the participants of the discussion. We hope to leverage our experience in advancing cross-cultural communication and to find partners in some of the other nuclear weapons states in order to deal with these difficulties.
Tractability: challenges of figuring out what actually works
One of the core motivations for this project simultaneously constitutes one of its primary challenges: For all we can tell, figuring out what actually works to reduce nuclear risks effectively is extremely difficult, never mind the challenges of finding some degree of common ground in terms of identifying room for improvement. Moreover, we simply face quite high uncertainty when evaluating the effectiveness of different interventions to impact policy more generally: Goal prioritisation requires that we weigh the importance of different goals, which is mired by ethical and empirical controversies; and the choice of interventions and strategies requires that we attribute policy changes to specific actions, which is difficult in a world marked by non-linear effects, interdependence of causes, and feedback loops[5]. We are devising this initiative in order to make headway on those epistemic and collaborative challenges, and we are driven by the belief that it constitutes a worthwhile attempt: while it certainly is not guaranteed to succeed, it does hold some promise in delivering improvements in the way we approach complexity[6].
Resource constraints
Our initiative is partially motivated by the urgency for effective action created by the shortage of funding, but we also realise that these resource constraints pose limitations and challenges to what a forum for bringing together NWPGs may achieve. If the challenges to communication and tractability can be met, there is still a chance that the insights generated by the forum cannot be put into practice because groups lack the finances to implement new strategies effectively. Although this is an issue, collective strategies that create efficiency savings and more stringently prioritise focused efforts will likely provide more financing per initiative. We also plan to work with nuclear policy funders to integrate their insights and communicate funding requirements and issues across the community.
Funding gaps also raise a further challenge for our efforts to get buy-in from relevant NWPGs: many of them may already be facing a backlog of tasks and projects they would like to implement, raising their reluctance to allocate some of their employees’ time to our initiative. It’s thus imperative that such an initiative has an explicit objective to save time, through shared knowledge acquisition, facilitating more efficient community discussion and avoiding duplication of efforts.
Possible adverse effects
Several of the challenges we described until now don’t just threaten the benefits of our initiative, they also create risks of causing inadvertent harm:
If done or framed in an inconsiderate way, our project might end up increasing competition and hostility between groups (if the initiative is perceived as a forum for ranking and then prioritising resource allocation between different groups and their strategies).
If done naively, rigorous impact evaluations could lead to a narrow-minded and thus inaccurate assessment of different strategies, which could engender unwise action recommendations for allocating funding, designing strategies, and implementing interventions.
If communicated naively, our emphasis on rigorous impact evaluations might reinforce unhelpful norms and behavioural pressures for civil society actors, creating the perceived and/or real need to spend significant resources on “selling” their activities to funders instead of engaging in activities that they believe will actually advance the goal of nuclear risk reduction.[7]
We take these possible negative side-effects extremely seriously and seek to prevent such harm by adopting a risk-aware and self-critical perspective throughout our work on the project. We also encourage participation from those that might harbour considerable scepticism or concern for the approach of our project. For this reason, comments and feedback on this last section would be particularly valuable to us!
“We”, that is: Paul Ingram and Sarah Weiler, in collaboration with others in the nuclear policy field. We have been collaborating for the last year informally to plan and pursue joint projects largely focused on nuclear winter awareness.
Aside from the common-sense case for such discussions, there are numerous studies that indicate that connecting diverse perspectives on tricky questions improves decision quality.
The fall in funding is primarily due to MacArthur’s withdrawal from the field, and is exacerbated by the implosion of FTX and the fall in value of Facebook shares, which reduces Open Philanthropy’s deployable capital.
This includes the breakdown of the START treaty, Russia’s deratification of CTBT, and military modernisation and build-up programmes in several nuclear weapons states.
More extensive discussions of the challenges of uncertainty in policy impact evaluations can be found in Justis 2022, Woods 2022, and Wiblin 2016 (to name just three examples). An early analysis that discusses those challenges specifically with regards to the nuclear policy field is given in Betts 1979.
For a more detailed justification of that belief, see the section “The basic case for coordination and synthesis” above.
A similar concern was recently raised in Elizabeth 2023, and is also touched on Karnofsky 2022 (see especially this section). The general problem is well-described by Goodhart’s law (see LessWrong for a quick intro).