Abstract
There are many cases in which people collectively cause some morally significant outcome (such as a harmful or beneficial outcome) but no individual act seems to make a difference. The problem in such cases is that it seems each person can argue, ‘it makes no difference whether or not I do X, so I have no reason to do it.’ The challenge is to say where this argument goes wrong. My approach begins from the observation that underlying the problem and motivating the typical responses to it is a standard, intuitive assumption. The assumption is that if an act will not make a difference with respect to an outcome, then it cannot play a significant, non-superfluous role in bringing that outcome about. In other words, helping to bring about an outcome requires making a difference. I argue that the key to solving the problem is to reject this assumption. I develop an account of what it is to help to bring about an outcome, where this does not require making a difference, and I use this explain our reasons for action in the problem cases. This account also yields an error theory that explains why the standard assumption is so tempting, even though it is mistaken.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Slightly adjusted from Parfit (1984, 76).
In previous work I have called it ‘the problem of collective harm', and the cases ‘collective harm cases', but this new terminology better captures the scope of the problem. I am grateful to Andrew Franklin-Hall for this suggestion.
In saying that the focus is on the reasons question rather than the obligations question, I don’t mean to suggest that just any reason will do. To solve the problem of collective impact we need a reason that has to do with, in a central way, the morally significant outcome in question. There may be many reasons not to fly to Europe for vacation this summer: to avoid the stress of peak-season travel, to have more money to donate to charity, to have more time to volunteer at the homeless shelter, and so on. But none of these reasons address what is at issue in the problem of collective impact that concerns individual decisions to fly and global warming. To solve that problem we need to give a morally relevant reason that has to do with the fact that air travel is a major contributor to global warming. So, to put the upshot here more precisely: the distinctive, core challenge that I’m interested in addressing is one of understanding how it is, in collective impact cases, that there is morally relevant reason to X having to do with morally significant outcome Y, when it looks as though X-ing won’t make a difference with respect to Y.
Throughout this paper when I say that an act ‘makes no difference’, this is short for ‘makes no difference with respect to the morally significant outcome in question.’ Of course, my act of adding a pint makes a difference to something; it makes a difference to how many water molecules are in the cart, for instance. But the point is that it makes no difference with respect to the morally significant outcome of concern: the relief of the men’s suffering.
Or at least, I will suppose that it does for the purposes of the discussion.
Note that in Nefsky (2011) I used different terminology for the same distinction. I called these ‘triggering cases' and ‘non-triggering cases'.
To be sure: my explication of this way of using the term is not an everyday-language explication; we don’t regularly use terms like ‘non-superfluous’ and ‘causal role’ in ordinary conversation. But the usage of ‘helping’ that it picks out is ordinary. Examples of it being used in this way include: ‘I need to study tonight; doing well in this course could help me get into med school’, or ‘Don’t forget to take your walk today: even just a 10 min walk will help to keep you in shape,’ or ‘don’t remove that plant; if you remove it, this could help erode the hillside.’ As the last example illustrates, on this usage, ‘helping’ is not specifically a ‘positive’ notion: it can be used for playing a non-superfluous role in bringing about good or bad outcomes. But when the outcome is not specified explicitly in the statement (e.g. ‘do X: it could help’) it is typically a good outcome that is implied.
This is not to say that the moral significance is built right into this concept of ‘helping’ itself. One could take a view of morality that denies it. If you somehow think that what one does, or tries to do, in the world has no bearing on the morality of one’s conduct (e.g. if you think that morality is just a matter of whether you are acting on the basis of consistent beliefs), then you wouldn’t think that the fact that one could help to bring about a good outcome, or help to prevent a bad one, would give any morally relevant reason to do so. But most of us do think that what one does or tries to do in the world matter morally, and we are correct to think so.
See Kagan (2011), Arntzenius and McCarthy (1997), Parfit (1988), and Norcross (2004). Otsuka (1991) and Regan (2000) also attempt to argue that there must be at least a chance of making a difference. For an example of a different ‘denying the description’ approach see Glover’s appeal to fractions of perceptible harm in Glover (1975).
See Nefsky (2011).
Mark Budolfson convincingly demonstrates that, contrary to those who try the expected utility approach, the expected goodness calculation often does not come out anywhere close to recommending the acts in question in threshold cases (Budolfson forthcoming).
See Nefsky (2015).
I say ‘primary’ reason because I don’t rule out that other considerations can provide additional support for acting. For instance, if we have an account like mine, the views in the second category (e.g. the appeal to fairness) can make sense, and so could perhaps provide additional reasons for action.
See the discussion of C7 in Parfit (1984).
Kutz (2000).
They are different usages but closely related: often one helps in the former sense (participates in a collective endeavor) by doing an act that helps in the latter sense (plays a significant part in bringing about an outcome).
Goldman (1999) takes such an approach. I think Goldman and I share a central intuition: that there can be reason for action having to do with the causal role your act could play, even if it won’t make a difference. But what Goldman—who is focused specifically on voting—takes this to mean is that what he needs to do is show that a vote can be part of the cause, even when it makes no difference. His paper is, thus, focused on rejecting the simple counterfactual model of causation and on looking for a model of causation that can capture an individual vote as causally involved. This, as I will argue in a moment, is not what needs to be done.
Compare this to a case in which I simply dump out my water supply upon realizing that the cart is already full. That would be a similarly wasteful act, but not one that is causally involved in bringing about that the men’s suffering is alleviated.
On Goldman’s use of the word ‘helping’ (see note 19 above) my act of power-hosing would count as helping, because ‘helping’ for him just is ‘part of the cause’. He is using the term differently than I am. Goldman proposes two different models of causal involvement: a version of Mackie’s INUS account and what he calls a ‘vectorial causal system’. Both models count my vote for Y as part of the cause of Y winning the election and my act of power-hosing as part of the cause of the alleviation of the men’s suffering, as they should if they are accounts of causal involvement. Neither does anything to explain how the vote can be non-superfluous, even though the power-hosing is superfluous. So, no light is shed on why one has reason to do the former but not the latter. A similar problem arises for Tuck (2008), who takes a similar approach to Goldman.
I defined helping as ‘playing a non-superfluous causal role’, but it is worth noting that the role actually need not be causal. I am focusing on causation for simplicity, but the account can easily be extended to apply to non-causal ‘bringing about’ relations that acts can stand into outcomes. In particular, we can extend the account to cover playing a non-superfluous constitutive role. We can do this by simply changing the supposition to: ‘Suppose that your act of X-ing could be part of what causes or constitutes Y.’
Strictly speaking it’s probably not true that more votes in favor of Mr. Powers are needed if Dr. Evil is to be stopped. Dr. Evil could be assassinated, for instance. What matters is not whether more acts of X-ing are needed, but whether (*) is satisfied. And (*) can be satisfied as long as there is at least one viable route to Dr. Evil’s defeat that involves acts of voting for Mr. Powers. This connects to the importance of the ‘at least in part’ in (*). It would be a mistake to think that for an act of X-ing to be non-superfluous with respect to Y, it must be possible that Y will fail to come about due solely to a lack of X-ing. First, there could be many sorts of things besides X-ing that need to happen for Y and implausible that the failure of Y would be attributable to a lack of any one of them alone. Moreover, as we just saw, there could be more than one viable route to an outcome only some of which involve X-ing. If Y does not occur, then the failure of Y won’t be due just to a lack of X-ing; it will be due partially to not enough X-ing and partially to a failure to take the other routes.
I have been focusing on examples of collective impact cases in which enough acts of a certain sort will together bring about a good outcome or prevent a bad one. But there are many cases in which it is rather that enough acts of a certain sort will collectively cause some bad outcome, and what we need is an explanation as to why each individual has reason to refrain from acting. How does my account handle such cases? The first thing to notice is that, just as you can help to bring about a good outcome or help to prevent a bad one, you can help to bring about a bad outcome or help to prevent a good one. Helping need not be a ‘positive’ thing. Whether it is or not depends on the outcome. We have said that if by acting in a certain way I might help to bring about a good outcome or help to prevent a bad one, this gives me reason to do so. Similarly, we can say that if by acting in a certain way I might help to bring about a bad outcome, or help to prevent a good one, this can give me reason to refrain from acting in that way. That is, if acting in a certain way might make a non-superfluous causal contribution toward bringing about what is now an avoidable bad outcome, this can give one reason not to do so.
That said, while my view does not imply or otherwise endorse it, it is consistent with my view that your being part of a collective is normatively significant, and so that there are additional reasons for action coming from the participatory character of your act. My claim is just that your potential to help provides the main—or, if you prefer, an important and central—reason for action in these cases. Now, as I mentioned in Sect. 2, Nefsky (2015) argues that views like participation views (views that fall in the second of the two categories of typical response) cannot actually work unless we can say that individual acts of the relevant type can do something causally significant in these cases. This appears to be a real problem for such views. But if that is indeed a problem for such views, then the account that I am developing in this paper solves that problem for them.
I should note that I am happy accepting that the fact that there is a miniscule chance of the election turning on your vote provides you with a reason to vote. What I deny is that this is the main or only reason to vote, and in particular, the main or only reason having to do with the potential instrumental significance of your individual vote.
Either that or they must bite the bullet and accept that there is no reason to give your pint to the cart having to do with the suffering men in the desert.
Nefsky (2011).
Nor is it strictly speaking necessary that more water is needed. See note 23 above.
Or, to put it the other way around: your paying your bill in full is not going to help to bring about the collapse of the system.
Does this mean that, on my account, if all the opinion polls are predicting that Powers will win by a large margin in an upcoming election, or that if you live in a non-swing-state, then your vote cannot help to elect him, and so that you don’t have that reason to vote? No; while people cannot pay their bills in full if they simply do not have the money to do so, it is not impossible for people to change their minds about who they will vote for or about whether to bother voting. But the reason to vote could be weaker in these situations than it would be if the election were more up in the air, or if you were living in swing state. I discuss this in Sect. 6, ‘the Notion of Possibility’.
What do I mean by ‘ordinary’ or ‘typical’ non-collective impact cases? The ‘non-typical’ cases I have in mind are not actually uncommon. Rather, they are non-typical only because they are analogous to collective impact cases, and so give rise to a largely analogous problem. These are cases in which, instead of a group of people doing acts of a certain sort, we have a single person doing acts of a certain sort over a period of time. By acting in this way over time, an outcome of significance (whether from a moral or prudential perspective) results. But, as in collective impact cases, no individual such act seems to itself make a difference, and thus it’s difficult to explain why there is reason to do it on any given occasion. For example, while smoking many cigarettes could lead to health problems, for each cigarette one could wonder: will it make any difference to my health, and if it won’t, why not enjoy it? A nice feature of my proposed solution is that it carries over as a solution to this analogous problem as well. For discussions of these cases, see Quinn (1990), and Tenenbaum and Raffman (2012). For discussions that specifically link collective impact cases and these cases, see Andreou (2006), Elster (1985), and Arntzenius and McCarthy (1997).
In collective impact cases, this is due to the large-scale nature of the case combined with epistemic barriers to knowing where the thresholds are and where things stand in relation to them. In the voting case, a huge number of votes are needed if our candidate is to win and, at the time when we must decide whether or not to vote, we cannot know what that threshold is or where things stand in relation to it. This makes it overwhelmingly unlikely that the threshold will be hit exactly.
‘Most likely will not’ in threshold cases, and ‘certainly will not’ in non-threshold cases.
It is logically and metaphysically possible that by clapping our hands we will alleviate the men’s suffering, but as far as the notion of possibility relevant in practical deliberative contexts is concerned, this is not possible.
For instance, something like Keith DeRose’s analysis of epistemic possibility could work well (DeRose 1991).
Conditional possibility, which is an account of which actions are possible for an agent, comes to mind here. According to this account, the actions that are possible for me are those that I would do, if I chose to do them. (For an argument in favor of this account see, for example, Bok 1998.) But this is just about which actions are possible for an agent. We need a general notion of possibility. Perhaps we can extend this notion of conditional possibility. Perhaps, for instance, we can say: Z is possible if and only if there is at least one set of choice-assignments (i.e. assign a choice to each relevant person) such that if these choices are made, then Z will come about. This might be a promising start, though further qualification will be needed. (In particular, one problem with conditional possibility that would carry over here is that there may be some actions that we should say you cannot choose even though if you did choose them, you would do them.) In any case, I am not at all committed to this proposal, and I leave it as an open question which notion of possibility works best.
The 2016 US presidential election was an example of this: polls leading up to the election had Hilarly Clinton in the lead, and yet she lost to Donald Trump. For another example, in the 2013 British Columbia general election, every public opinion poll prior to the election predicted that the NDP candidate would win by a wide margin; despite that, the Liberal candidate won.
It seems to me that, on the notion of possibility that we employ in deliberation, we do recognize it as very remotely possible that he will: we recognize that some bizarre (but not impossible) turn of events might result in Obama’s dropping by for tea. However, because this possibility is so remote, we do not give it any weight in deliberation, and can correctly count it as impossible for all practical purposes.
Within the context of practical and moral deliberation, we usually treat the question of whether people will choose to X as at least as open as it is within their own deliberations. And it seems to me that we are correct to do so. Indeed, we should really treat it as more open than that, since it is not uncommon for people to reopen deliberative questions that they previously considered closed. Since very many people deliberate about whether to act in the relevant ways in collective impact cases, and since others still could open or reopen the question, we generally should not consider it a closed question what they will do.
If it is something much more substantial that is at stake, then it could be that the helping-without-making-a-difference consideration is outweighed. But this seems like a good thing, rather than a worry. If, say, the only way for me to vote in the national election involves incurring a high risk of losing my job and having to struggle to support my family, then it is plausible that the considerations in favor of voting are outweighed. A view that yields that you must vote no matter what, at all costs, despite the fact that it’s extremely unlikely that your vote could itself make a difference, is not plausible.
This includes Kantian approaches, approaches that appeal to fairness, and approaches that appeal to participation in a collective action. See Nefsky (2015).
References
Andreou, C. (2006). Environmental damage and the puzzle of the self-torturer. Philosophy & Public Affairs, 34, 95–108.
Arntzenius, F., & McCarthy, D. (1997). Self torture and group beneficence. Erkenntnis, 47, 127–144.
Bok, H. (1998). Freedom and responsibility. Princeton: Princeton University Press.
Budolfson, M. (forthcoming). The inefficacy objection to consequentialism and the problem with the expected consequence response. Philosophical Studies.
Cullity, G. (2000). Pooled beneficence. In M. Almeida (Ed.), Imperceptible harms and benefits (pp. 1–23). Dordrecht: Kluwer Academic Publishers.
DeRose, K. (1991). Epistemic possibilities. Philosophical Review, 100(4), 581–605.
Elster, J. (1985). Rationality, morality, and collective action. Ethics, 96, 136–155.
Glover, J. (1975). It makes no difference whether or not i do it. Proceedings of the Aristotelian Society, 49, 171–190.
Goldman, A. I. (1999). Why citizens should vote: A causal responsibility approach. Social Philosophy & Policy, 16(2), 201–217.
Jackson, F. (1987). Group morality. In P. Pettit, et al. (Eds.), Metaphysics and morality: Essays in honour of J.J.C. Smart (pp. 91–110). Oxford: Blackwell.
Kagan, S. (2011). Do I make a difference? Philosophy & Public Affairs, 39(2), 105–141.
Kutz, C. (2000). Complicity. Cambridge: Cambridge University Press.
Nefsky, J. (2011). Consequentialism and the problem of collective harm: A reply to Kagan. Philosophy & Public Affairs, 39, 364–395.
Nefsky, J. (2015). Fairness, participation, and the real problem of collective harm. In M. Timmons (Ed.), Oxford studies in normative ethics (Vol. 5). Oxford: Oxford University Press.
Norcross, A. (2004). Puppies, pigs, and people: Eating meat and marginal cases. Philosophical Perspectives, 18, 229–245.
Otsuka, M. (1991). The paradox of group beneficence. Philosophy & Public Affairs, 20, 132–149.
Parfit, D. (1984). Reasons and persons. Oxford: Oxford University Press.
Parfit, D. (1988). What we together do (Unpublished draft).
Quinn, W. (1990). The paradox of the self-torturer. Philosophical Studies, 59, 79–90.
Regan, D. (2000). Perceiving imperceptible harms: With other thoughts on transitivity, cumulative effects, and consequentialism. In M. Almeida (Ed.), Imperceptible harms and benefits (pp. 49–73). Dordrecht: Kluwer Academic Publishers.
Sinnott-Armstrong, W. (2005). It’s not my fault: Global warming and individual moral obligations. Perspectives on Climate Change: Science, Economics, Politics, Ethics, 5, 293–315.
Strang, C. (1960). What if everyone did that? Durham University Journal, 53, 5–10.
Tännsjö, T. (1989). The morality of collective action. The Philosophical Quarterly, 39, 221–228.
Tenenbaum, S., & Raffman, D. (2012). Vague projects and the puzzle of the self-torturer. Ethics, 123(1), 86–112.
Tuck, R. (2008). Free riding. Cambridge: Harvard University Press.
Acknowledgements
I owe special thanks to Niko Kolodny, Alex Rennet, Sergio Tenenbaum, and R. Jay Wallace for extensive discussion of and comments on earlier drafts of this work. I am also very grateful to Joe Campbell, Candice Delmas, Thomas Dougherty, Luke Elson, Elizabeth Harman, Tom Hurka, Ewan Kingston, Markus Kohl, Melissa Lane, Mark LeBar, Alex Madva, Parisa Moosavi, Andrew Newton, Douglas Portmore, Hamish Russell, Carolina Sartorio, Walter Sinnott-Armstrong, Holly Smith, Tatjana Visak, Henry West, Seth Yalcin, and an anonymous reviewer for Philosophical Studies for helpful comments. Finally, I would like to thank the participants in the 2014 Colloquium in Legal, Political and Social Philosophy at NYU; audiences at Berkeley, McMaster University, University of Toronto, University of Waterloo, the 2013 Rocky Mountain Ethics Congress, and the 2015 APA Pacific Division meeting; and the participants of my 2012 graduate seminar at the University of Toronto.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Nefsky, J. How you can help, without making a difference. Philos Stud 174, 2743–2767 (2017). https://doi.org/10.1007/s11098-016-0808-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-016-0808-y