• Christian Barry and Katie Steele

Individual Difference-Making and Climate Change: Tragedy of Collective Action or Culpable Collusion?

There may be cases where people collectively cause a morally significant outcome, but no individual act seems to make a difference, or at least enough of a difference to give them sufficient reason not to act as they do. Climate change is often portrayed as a case of this sort—a collective action problem of the grandest scale. The idea is that, holding what others do fixed, no individual agent has very weighty reason to reduce emissions, since any such behavioural change would not yield sufficient expected harm reduction to outweigh the costs to the agent. Consequently, no one of these individuals is accountable for the harms resulting from climate change (Kingston and Sinnot-Armstrong 2018). And yet the tragedy is that the simultaneous emissions reduction of many agents, yielding a large aggregate harm reduction, would easily warrant the costs borne by each of them to achieve this aim.

To clarify, the tragic-collective-action model of climate change does not depend on individuals making no significant difference at all to the expectation of harms due to climate change. Rather, it relies on the fact that, conditional on what others do, no individual makes enough expected difference to make it the case that they are morally required to bear the costs of reducing their emissions.

The claim that individuals are not making enough expected difference to climate-change related harms through their individual lifetime emissions has been contested (Broome 2012, 2019, Nolt 2011). But let us assume that this is the case – that individual actions to reduce emissions, when taken independently of others, impose costs on agents that exceed what they are morally required to bear for this purpose. Nonetheless, the question remains whether this is the right characterisation of the problem of climate change – why action on climate change does not and perhaps will not occur. To describe the problem of climate change in this way is to suggest a relatively ‘innocent’ explanation: while sufficiently many people care appropriately about climate change, each rightly judges their own contribution to solving the problem, taken independently of others, to be too trivial to make their conduct wrong. Call this the innocent collective action explanation. Here, however, we will consider a much less innocent model of the problem that, we think, may actually align better with agents’ actual preferences, in particular, models in which agents collude in the generation of harm, which we’ll call the culpable collective collusion explanation.

The contrast between innocent collective action and culpable collective collusion explanations can be illuminated through an adaptation of Derek Parfit’s Harmless Torturers case (Parfit 1984, 80) to the twin cases of Innocent Torturers and Culpable Torturers as follows: In both cases, many individuals administer a very tiny pain via electric shock to a victim such that the aggregate effect is a very severe pain for the victim (for simplicity, let’s assume, in contrast to Parfit’s original case, that the difference each shock makes is perceptible, but slight). In Innocent Torturers, the situation is such that each of those causing the shocks is using some device that will cure their child of a mild headache. They each produce an electric shock yielding a tiny increase in pain for the victim as an unavoidable side effect. Holding fixed what others do, each judges, rightly (let’s say), that their own child’s cured headache can justify the increase in pain that they thereby inflict on the victim. However, in aggregate, the many cured headaches are not worth the resulting very severe pain of the victim (let’s say), and the participants are each sensitive to this moral fact. In Culpable Torturers, by contrast, the individuals involved have colluded in each administering shocks because they take delight in the victim experiencing severe pain and yet none of them are able to deliver this magnitude of pain single-handedly and so they must cooperate. (To make the cases objectively identical, one can further assume that a mere side effect of the severe pain, from the point of view of the participants, is that many children are relieved of headaches.) Note that in this case the fact that no-one of the individuals involved in the administration of the shocks makes much of a difference to the pain experienced by the victim does not shield them from accountability for the extreme pain he suffers. Indeed, if imposing high costs on each of the contributors in this case were a necessary means of protecting the victim from the harm they collectively cause him, it would seem permissible to do so.

While individual agents may not be so morally perverse as to seek climate change for its own sake, they may nonetheless inappropriately seek the benefits that a world with climate change brings them.

Innocent Torturers and Culpable Torturers have the same structure if we restrict ourselves to considering only the role of each person’s administration of a shock, independently of those of others, and the resulting harm (increased pain) and benefit (headache relief). And yet ‘the problem’ in each case – the reason why the shocks continue when they ought to be stopped – seems very different given the further details that distinguish the cases. We will return to why the further details matter shortly.

First, what reason is there at all to think that the problem of climate change – why climate change continues unabated and why an alternative path of mitigation is difficult to enact – may resemble Culpable Torturers? After all, it is not clear that any agents take delight in the onset of climate change. And surely there are no secret conspiratorial meetings about how to bring such harm about. Still, the intentions and cooperative networks of agents with respect to climate change may well still be describable as involving collusion. Take first the intentions of the agents involved. While individual agents may not be so morally perverse as to seek climate change for its own sake, they may nonetheless inappropriately seek the benefits that a world with climate change brings them. That is, they may seek the benefits associated with many other agents continuing with business as usual and not reducing their carbon emissions, and not merely the benefits associated with not reducing their own individual carbon emissions. So indirectly, they may indeed seek cooperation in causing climate change, just as the agents in the Culpable Torturers seek cooperation in causing extreme pain to the victim. The principal difference would be that instead of valuing the aggregate harm of climate change per se, they would merely fail to disvalue this harm enough to cancel out the benefits to them of the activities causing aggregate climate change. Now consider the conspiratorial aspect. While there may be no meetings where individuals get together and say “what can we do to bring about climate change?”, there may well be plenty of ways that these individuals signal to others that they are keen to support the prevention of actions that would mitigate climate change and would thus impose costs on them. In effect, they may more or less surreptitiously support norms that bolster the status quo of unabated climate change. This can be true even of those that continue to actively send signals through the market that they will continue to welcome the benefits that business as usual will bring, even if they claim to be concerned about climate change.

So even granted the assumptions mentioned earlier about certain aspects of the causal and moral structure of climate change, the problem of responding to climate change, or the explanation of our continuing inaction, may plausibly take very different forms: it may be principally one of innocent collective action or rather principally one of culpable collective collusion.

While there may be no meetings where individuals get together and say “what can we do to bring about climate change?”, there may well be plenty of ways that these individuals signal to others that they are keen to support the prevention of actions that would mitigate climate change and would thus impose costs on them.

How do these explanations of the problem really differ and which seems more adequate? Consider that, although collective action problems do indeed arise in public life, there are many optimistic stories about our abilities to solve them, even without a central overarching government that can alter the relevant incentives, so long as we generally want the same thing and where the only obstacle is that we must trust that our actions will be matched by like-minded others. The Noble Laureate Elinor Ostrom has documented the emergence of such cooperation in her work, for example, in the development of irrigation institutions in Spain and the Philippines (Ostrom 1990). Indeed, in a case like Innocent Torturers many would find it puzzling if the individuals involved were not able to coordinate their actions to prevent the severe pain of the victim, once they became aware of the facts of the scenario. In effect, once participants become aware of the situation in innocent collective action scenarios, there is scope for more difference-making per unit cost on the part of each individual than what one might have first surmised: all involved have mutual influence over each-others’ behaviour and can exploit common desires to participate in advantageous social norms. By way of solution, each might signal their willingness, conditional on the participation of others, to refrain from, say, high emissions activities or the administering of a shock, thereby beginning a process by which a stable social norm against such behaviour could emerge. Calls for a ‘green recovery’ post-COVID can be seen as attempts at such signaling.

That is not to say that the establishment of harm-mitigating social norms is inevitable or that it will always be easy, especially where numerous actors are involved and there are multiple ways to distribute the costs of harm reduction, each favouring some actors over others (Ostrom 1990, Barrett 2007). But like-minded agents can develop strategies to overcome these obstacles, such as democratic participation in determining the distribution of costs and adequate monitoring to expose free-riding. So there is reason to think that if the problem of climate change really were principally one of collective action, then the individuals involved would have made more progress than we have in developing stable norms to address it.

Problems of the form culpable collective collusion do not admit of the same kind of solution, since the agents involved do not have sufficient will to participate in harm reduction to start with. Worse, they actively influence each other in preventing the establishment of harm-reducing norms. (In this way, one could say that they have in fact solved their own morally perverse collective action problem.) If climate change were principally a case of culpable collective collusion, then it would be no surprise that efforts to establish social norms that presume all would prefer effective collective action on climate change (but are merely doubtful about the role of their own particular contribution) would be unsuccessful. Such efforts would be as futile as trying to persuade the agents in Culpable Torturers to refrain from submitting their shock because in so doing they will raise the probability that others will refrain too. That is precisely not what they want.

[W]e are suggesting that the best explanation of inaction on climate change seems likely to be that the forces of collusion overwhelm the forces of cooperation for the good.

That there is some collusion to block efforts at climate change is of course a familiar thought. The question is the extent of this collusion. Normally, the actors that are represented as being involved in such collusion are fossil fuel companies, and perhaps governments that are unduly beholden to them. That there is such collusion seems evident. But it is not clear that such collusion is limited to these agents or else that their power is duly limited by the voting public. Indeed, we are suggesting that the best explanation of inaction on climate change seems likely to be that the forces of collusion overwhelm the forces of cooperation for the good. Either fossil fuel companies and the like are even more powerful than one might have thought, or else many other actors too, including the majority of ordinary individual people, are involved in supporting the collusion effort, even if they do not regard climate change itself as a goal. Insofar as individuals are not signalling their willingness to cooperate in the establishment of social norms that better serve to mitigate climate change and are instead signalling their willingness to cooperate in prevailing norms that accelerate it, they too seem to collude in the production of the resulting harms.

Looking at the problem of climate change as a case of very widespread collusion calls for a change in focus: the fact that any agent’s contributions to reducing carbon emissions, taken independently, would be negligible, is neither here nor there. The problem is that sufficiently many actors collude with or influence each other to prevent rather than obtain a collective mitigation effort. The influence of these ‘bad actors’ therefore needs to be countered, either by changing their mind, buying them off, or imposing severe costs on them should this be necessary to compel them to act otherwise.


Barrett, S. 2007. Why Cooperate? The Incentive to Supply Global Public Goods. Oxford, UK: Oxford University Press.

Broome, J., 2012. Climate Matters: Ethics in a Warming World. New York: Norton.

Broome, J., 2019. ‘Against Denialism’, The Monist 102: 110–29.

Kingston, E., and Sinnott-Armstrong, W., 2018. ‘What’s Wrong with Joyguzzling?’, Ethical Theory and Moral Practice, 2:169–86.

Nolt, J., 2011. ‘How Harmful Are the Average American’s Greenhouse Gas Emissions?’, Ethics, Policy and Environment14: 3–10.

Ostrom, E., 1990. Governing the Commons. Cambridge: Cambridge University Press.

Parfit, D., 1984. Reasons and Persons. Oxford University Press.

Disclaimer: Any views or opinions expressed on The Public Ethics Blog are solely those of the post author(s) and not The Stockholm Centre for the Ethics of War and Peace, Stockholm University, the Wallenberg Foundation, or the staff of those organisations.

  • Facebook
  • Twitter

©2020 by Public Ethics.


The views expressed in these posts are those of the author(s), and do not necessarily reflect the views of the Public Ethics blog or associated organisations.