Julia Nefsky is an Assistant Professor in the Department of Philosophy at The University of Toronto. Before that, she was a graduate student at UC Berkeley. She works in ethics. Her main current research project – which is the subject of this post – attempts to understand individual morality in contexts of collective impact. She is also working on some related questions about individual rational agency over time.

The Problem of Collective Impact

Julia Nefsky

Thank you so much to Meena for this terrific series, and for inviting me to contribute.

Much of my current research aims at understanding our individual moral reasons and responsibilities in contexts of “collective impact.” These are contexts in which if enough people act in certain ways rather than others, certain harms or injustices will be avoided or reduced, and yet no individual such act seems to make a difference. Climate change, for instance, is caused by millions of people driving, flying, using air-conditioning, and so on, but give or take any one such act and things will surely be just as bad. Or, consider that while a large-scale shift in demand away from factory-farmed meat would result in many fewer animals being tortured on factory-farms, it is doubtful that any one purchase makes a difference: the factory-farm is not going to be making different production decisions give or take any one purchase. The core problem in such cases is that if it’s true that acting in the relevant way won’t make a difference, it’s unclear how we can say that anyone ought, or even has reason, to do so. Each person can ask, “my doing this won’t make a difference, so what is the point?” Call this, “the problem of collective impact.”[1]

I argue that the key to solving this problem is to reject a basic assumption that underlies it. The assumption is that to make instrumental progress toward an outcome – or, as I often put it, to help to bring about an outcome – one must be able to make a difference with respect to that outcome. If acting in a certain way won’t make a difference, this means that things will be the same with respect to the outcome of concern whether or not you do so. And what we take this to mean is that acting in this way would be instrumentally useless – that it would not actually do anything significant toward bringing about that outcome. While this assumption is quite entrenched, I think it is false, and I argue that seeing this is a crucial and central component of solving the problem.

On why I think it is a crucial component of solving the problem: In Fairness, Participation and the Real Problem of Collective Harm, I argue that it is doubtful that we can solve the problem unless we can refute the impression that an individual act is instrumentally superfluous in these cases. Views that might seem to bypass having to worry about such a thing, cannot actually do so if they are to address what is at issue. (This includes various non-consequentialist theories, like Kantianism). (I give further support for this in Individual Consumption and Collective Impact.) However, at the same time, I don’t think we can in general deny that an individual act won’t make a difference in these cases. First, in Consequentialism and the Problem of Collective Harm, I argue that there are cases in which one’s act simply has no chance of making a difference. Second, even when there is a chance of making a difference, this chance can be miniscule – to the point where we can say that such an act, for all practical purposes, won’t make a difference. For this second point, see Mark Budolfson’s excellent paper, The Inefficacy Objection to Consequentialism and the Problem with the Expected Consequences Response. If these conclusions are correct, they together show that to solve the problem we need to reject that basic assumption: we need to show that an act can make significant instrumental progress even if it won’t make a difference.

While we do typically equate not making a difference with being instrumentally useless, I think that the idea that these two notions can come apart is also implicitly present in commonsense thought. To illustrate, take a classic collective impact case: voting in a large election. Suppose the upcoming US election ends up being between Trump and Clinton. In this case, many people will vote for Clinton because they want Clinton to be elected rather than Trump, and because they think that casting their ballot for Clinton does something toward that goal. That is, for many Clinton voters, they are not thinking of their vote as instrumentally superfluous. Rather, it is precisely the opposite thought that moves them to vote: the thought that their vote could make some (small but real) progress toward getting Clinton into the White House. Now, when a person is thinking in this way, I don’t think that what is going on is that she is under the mistaken impression that her vote is likely to make a difference to whether Clinton gets elected. Thoughts about the winner being different depending on her vote are not – I conjecture – coming into it. If this is right, she must have another notion of instrumental progress implicitly in mind. This is not to say she has any clear, consistent understanding of this notion. Indeed, when someone presses her on the fact that her vote most certainly won’t make a difference, she is likely to respond by calling up other sorts of justifications for voting (e.g. that voting expresses support for Clinton, or that it is her duty as a democratic citizen.) The point is just that people sometimes do think of their vote as making real instrumental progress toward an outcome, without having any impression of it as making a difference with respect to that outcome. My project aims to show that they are getting something importantly right when they think in that way.

In How You Can Help, Without Making a Difference I advance the idea that an act can help in collective impact cases, even if it won’t make a difference. If I’m right about this, this vindicates an appealing, commonsense answer to why there is reason for action in these cases, namely: doing so could help. Importantly, while I argue it is a mistake to think that helping requires making a difference, I also argue that it would be a mistake to think that just any causal contribution helps. There is a distinction between causal contributions that make progress and those that don’t – between superfluous and non-superfluous causal involvement. This distinction is at the heart of what is difficult about the cases: what is difficult is that even if I am in the causal fray, my involvement seems entirely superfluous. I don’t think we can adequately understand collective impact cases if we collapse this distinction. This paper, thus, aims to show how we can draw this distinction along something other than difference-making lines.

In my work I draw attention to the fact that the “it won’t make a difference” claim challenges not only the thought that one is obligated to act in the relevant ways, but also – and more basically – the thought that there is any moral reason to do so. But this doesn’t mean that we should be satisfied once we have an answer to that more basic challenge. Plausibly, morality should yield moral obligations to act in the relevant ways, not just some consideration in favour of doing so. The next stage of my project, thus, aims to address the question of what our moral obligations look like in these cases.

In an in-progress paper, I am arguing that we are indeed obligated to act in the relevant ways, but not in the sense that most philosophers who have written about the topic seem to be looking for. I argue that our moral obligations in collective impact cases are – to borrow a term from Kant – imperfect. If, for instance, you are in a position to make choices that will help to mitigate climate change, this does yield an obligation to do so, but – I argue – given the way in which your individual choices can help (namely, without making a difference), this obligation is best understood as imperfect. What it amounts to is an obligation to refrain from flying, driving, and so on enough of the time, rather than an obligation (even a pro tanto one) to do these things at every moment that one can.

Thus, the view I am defending has an important point of agreement with philosophers who have ‘bitten the bullet’ on the obligations question. Walter Sinnott-Armstrong, for instance, argues that if you look at a particular act, like driving a gas-guzzling SUV on “this particular sunny Sunday afternoon”, it is not wrong. I agree with that. But what I argue is that this doesn’t mean that you don’t have an obligation to refrain from driving in light of climate change: the obligation is just an imperfect one. Whether you satisfy it can only be determined by looking at the choices you make over time. Isolating a particular action in a collective impact case, and trying to show that it is obligatory is usually a mistake.

[1] In past work I have called it “the problem of collective harm”. I now think “the problem of collective impact” is a better term for it.