Ex Ante and Ex Post Contractualism: A Synthesis


According to contractualist theories in ethics, whether an action is wrong is determined by whether it could be justified to others on grounds no one could reasonably reject. Contractualists then think that reasonable rejectability of principles depends on the strength of the personal objections individuals can make to them. There is, however, a deep disagreement between contractualists concerning from which temporal perspective the relevant objections to different principles are to be made. Are they to be made on the basis of the prospects the principles give to different individuals ex ante or on the basis of the outcomes of the principles ex post? Both answers have been found to be problematic. The ex ante views make irrelevant information about personal identity morally significant and lead to objectionable ex ante rules, whereas ex post views lead to counterintuitive results in the so-called different harm and social risk imposition cases. The aim of this article is to provide a new synthesis of these views that can avoid the problems of the previous alternatives. I call the proposal ‘risk-acknowledging’ ex post contractualism. The crux of the view is to take into account in the comparisons of different objections both the realised harms and the risks under which individuals have to live.

This is a preview of subscription content, log in to check access.


  1. 1.

    See Scanlon (1998). Hereafter all unattributed references are to this work.

  2. 2.

    For a discussion, see, e.g., Ridge (2001) and Parfit (2003).

  3. 3.

    Other defenders of ex post contractualism include, e.g., Sophia Reibetanz Moreau (1998), Michael Otsuka (2015), Sune Holm (2018), and Korbinian Rüger (2018).

  4. 4.

    For defences, see, e.g., Rahul Kumar (1999), Aaron James (2012), and Johann Frick (2015). Scanlon (2013) himself has also more recently endorsed the view.

  5. 5.

    In addition to Reibetanz Moreau (1998, 304), for explanations of this argument see also Frick (2015, 184–185), Fleurbaey and Voorhoeve (2013), Holm (2018, 233–234), and Otsuka (2015).

  6. 6.

    For discussions of this problem, see also, e.g., Ashford (2004, Sect. 4), Frick (2015, 201–203), and Kumar (2015, Sect. 3).

  7. 7.

    There are also other objections to ex ante contractualism. For example, it can be argued that the view makes the distinction between unidentified doomed children and identified doomed children morally significant in an objectionable way. See Rüger (2018, 245–251).

  8. 8.

    I set the previously much-discussed same harm cases in which we need to decide between saving one and saving many aside for reasons of space. Scanlon (1998, 232) originally tried to deal with these cases with the ‘tie break argument’ according to which each member of the group has an additional personal objection based on the fact, if the group is not saved, their presence makes no moral difference. This response has been both criticised (Otsuka 2001) and defended (Hirose 2001). The risk-acknowledging version of ex post contractualism introduced below can deal with the same harm cases in the same way as the different harm cases without needing to rely on the tie break argument.

  9. 9.

    For discussions, see e.g., Norcross (2002), Parfit (2003, Sect. 3) and Hooker (2003, 72–75).

  10. 10.

    Before Frick, many others too had argued that ex post contractualism threatens to be too confining by not permitting many intuitively permissible socially productive activities such as aviation and bridge building (Ashford 2004, 298–300; Fried 2012; James 2012, Sect. 2; Kumar 2015, Sect. 4; Lenman 2008, Sect. 12). The examples used and the problems they pose are structurally identical to Frick’s case and so I won’t discuss these arguments separately. Scanlon himself tried to deal with those case by relying on the idea of reasonable level of precautions (209; for a critical discussion, see Ashford (2004, 299) and Lenman (2008, 114)). The versions of risk-acknowledging ex post contractualism introduced below can deal with those cases exactly in the same way as with Frick’s case (see fn. 20 below).

  11. 11.

    Frick also argues that ex post contractualism cannot distinguish between the objections, which the luckless children have in the previous case, and the objections which the doomed children have in the cases in certain identified group of children will die because the relevant vaccine is not effective for them (Frick 2015, 200; see also Rüger 2018, 241–242). The problem is that the latter children have intuitively a more serious complaint to the policies that will disadvantage them.

  12. 12.

    There are also a number of mixed strategies according to which justifiability to each person is a function of both ex ante and ex post objections which individuals can make. For different versions of these hybrid views and their problems, see Fried (2012, Sect. 3).

  13. 13.

    For a response defending the Argument from Certain Loss, see Otsuka (2015) and Holm (2018, Sects. 4–5).

  14. 14.

    Broadly speaking, in addition to Frick, the first strategy has been pursued by James (2012) and Kumar (2015). Likewise, the second strategy has been pursued in different ways by Otsuka (2015) and Holm (2018).

  15. 15.

    For explanations of Scanlon’s view of the relevant objections, see, e.g., Ashford (2004, 277–279), James (2012, 267) and Kumar (2015, 41–42).

  16. 16.

    Lenman (2008, Sect. 13) suggests a different synthesis of ex ante and ex post contractualism. Lenman’s proposal is that, even if objections are to be evaluated from ex post perspective, even from this perspective we are to consider whether you can object to what someone did to you when that action is understood from the agent’s ex ante epistemic perspective. Understood in this way, we can discount many resulting harms as objections by their improbability. This differs the view outlined below as that proposal does not discount any realised harms by their improbability, but rather it considers living under a risk as a ground for additional objection ex post.

  17. 17.

    It could be argued that this makes it ever so slightly more wrong to shoot someone at point blank compared to playing Russian roulette on them when this leads to death. The main objection both victims can make is that they have been killed. However, both can also object to the fact that they were put under a serious risk. In the case of shooting someone at point blank, this risk is more serious given that the likelihood of the other person dying is higher than in the Russian roulette case.

  18. 18.

    See also Lenman (2008, 108), Fried (2012, 58) and James (2012, 282).

  19. 19.

    See also Ashford (2004, 278), Lenman (2008, 108 and Sect. 14), Fried (2012, 58) and Kumar (2015, 40–41).

  20. 20.

    Note that this case is the one described by Scanlon (239) and discussed in the very beginning of Sect. 3 on page 9. This case has two important features. Firstly, the consequences of both actions in the relevant situation are fully known and, secondly, if we save the one individual from death then no harm is caused to the members of the group whereas if we save the group then no harm is caused to the one individual. The social risk imposition case Mass Vaccination (Unknown Victims) (see Sect. 3 above) differs from this case in these two respects. For a discussion of that case, see below.

  21. 21.

    Here I follow Kumar (2015, 43–44). Below, I consider only the base-line risks which we all have qua members of the general population. As Kumar points out (ibid., especially fn. 27), different individuals can, of course, also have additional risks qua members of more specific reference classes. Yet, taking these additional risks into account would not change the conclusions below substantially (but see Kumar (2015, 45–47) for a discussion. This additional feature of the risk-acknowledging forms of ex post contractualism also enables these views to recognise the difference between the luckless children in Mass Vaccination (Unknown Victims) and the doomed children in the slightly modified version in which certain identified group of children will die because the Vaccine 2 is not effective for them (see Frick (2015, 200), Rüger (2018, 241–242) and fn. 11 above). The defenders of these views can claim that, even if both children die in the corresponding cases, a more serious risk is imposed on the latter and so intrapersonally aggregated objections which the doomed children have to Vaccine 2 are stronger than the corresponding objections of the luckless children.

  22. 22.

    Others who have accepted the relevant probabilities in terms of a frequency measure in this debate include Fried (2012, 50) and Kumar (2015, 43–44, including fn. 26 on other alternatives).

  23. 23.

    In fact, as the table above shows, Stephen himself has the same objection to Vaccine 1 as Ben given that both suffer the same harm of paralysis of a leg and risk p under that principle. This means that Stephen’s own objection to Vaccine 1 is stronger on this view than his own objection to Vaccine 2.

  24. 24.

    The objection to ex post contractualism according to which the view is too confining by ruling out socially productive activities such as aviation can be responded to in the same way (see fn. 9 above). First, it can be argued that these activities too save lives and prevent other serious burdens (Holm 2018, 241; James 2012, 272). It can then be argued that living under principles that forbid these activities both causes those very harms to some individuals and also imposes a more serious risk of harms to them too. Given that the rules that permit these activities cause similar burdens but not as serious risks to anyone (Kumar 2015, 48), these principles cannot be reasonably rejected.

  25. 25.

    Here my response follows Kumar (2015, 36–37).

  26. 26.

    In Lenman’s (2008, 116) words, at this point we might then ‘think of these approaches as complementary rather than necessarily competing’. However, the next section argues that the two approaches in fact lead to different conclusions in some cases.

  27. 27.

    As Rüger (2018, 255–256) correctly points out, Frick is wrong to think that ex post contractualism is any worse off at this point than his own ex ante view. Frick thinks that ex ante contractualists can avoid the problem here by relying on other wrong-making features of actions than the ones based on what we owe to others. Yet, if the ex ante contractualists are allowed to do this, then surely ex post contractualists too can rely on a similar pluralist strategy in the Miners (1 vs. 100) to explain why the identified individual should be saved in them.

  28. 28.

    On this point, see (236).

  29. 29.

    Here we need to recall that the principle that is to govern this case is to govern also all other cases in which we must compare whether to save an identified individual in a costly way or use the resources to take preventive measures that will save a large number of lives in the future. As Frick (2015, 178–179) points out, we cannot assume that any individual would be a victim of the relevant future accidents many times. Yet, living under the principles that use all resources to saving identified victims rather than taking any preventive measures will be risky because the risks imposed by all choices of this type will aggregative intrapersonally. Under these principles, any individual will face many situations that are dangerous because it’s never the case that the society takes the required steps to prevent accidents. For a discussion, see Rüger (2018, 254).


  1. Ashford, Elizabeth. 2004. Demandingness of Scanlon’s contractualism’. Ethics 113: 273–302.

    Article  Google Scholar 

  2. Fleurbaey, Marc, and Alex Voorhoeve. 2013. Decide as you would with full information! An argument against ex ante Pareto. In Inequalities in health: Concepts, measures and ethics, ed. Nir Eyal, Samia Hurst, Ole Norheim, and Dan Wikler, 113–138. Oxford: Oxford University Press.

    Google Scholar 

  3. Frick, Johann. 2015. Contractualism and social risk. Philosophy & Public Affairs 43: 176–223.

    Article  Google Scholar 

  4. Fried, Barbara. 2012. Can contractualism save us from aggregation? The Journal of Ethics 16: 39–66.

    Article  Google Scholar 

  5. Hirose, Iwao. 2001. Saving the greater number without combining the claims. Analysis 61: 341–342.

    Article  Google Scholar 

  6. Holm, Sune. 2018. The luckless and the doomed. contractualism and justified risk-imposition. Ethical Theory and Moral Practice 21: 231–244.

    Article  Google Scholar 

  7. Hooker, Brad. 2003. Contractualism, spare wheel, aggregation. In Scanlon and contractualism, ed. Matt Mattravers, 53–76. London: Frank Cass.

    Google Scholar 

  8. James, Aaron. 2012. Contractualism’s (not so) slippery slope. Legal Theory 18: 263–292.

    Article  Google Scholar 

  9. Kumar, Rahul. 1999. Defending the moral moderate: contractualism and common sense. Philosophy & Public Affairs 28: 275–309.

    Article  Google Scholar 

  10. Kumar, Rahul. 2015. Risking and wronging. Philosophy & Public Affairs 43: 27–49.

    Article  Google Scholar 

  11. Lenman, James. 2008. Contractualism and risk imposition. Politics, Philosophy & Economics 7: 99–122.

    Article  Google Scholar 

  12. Norcross, Alastair. 2002. Contractualism and aggregation. Social Theory and Practice 28: 303–314.

    Article  Google Scholar 

  13. Otsuka, Michael. 2001. Scanlon and the claims of the many versus the one. Analysis 60: 288–293.

    Article  Google Scholar 

  14. Otsuka, Michael. 2015. Risking life and limb. In Identified versus statistical lives: An interdisciplinary perspective, ed. Glenn Cohen, Norman Daniels, and Nir Eyal, 77–93. New York: Oxford University Press.

    Google Scholar 

  15. Parfit, Derek. 2003. Justifiability to each person. Ratio 16: 368–390.

    Article  Google Scholar 

  16. Reibetanz Moreau, Sophia. 1998. Contractualism and aggregation. Ethics 108: 296–311.

    Article  Google Scholar 

  17. Ridge, Michael. 2001. Saving Scanlon: Contractualism and agent-relativity. The Journal of Political Philosophy 9: 472–481.

    Article  Google Scholar 

  18. Rüger, Korbinian. 2018. On ex ante contractualism. Journal of Ethics and Social Philosophy 13: 240–258.

    Article  Google Scholar 

  19. Scanlon, T.M. 1982. Contractualism and utilitarianism. In Utilitarianism and beyond, ed. Amartya Sen and Bernard Williams, 103–128. Cambridge: Cambridge University Press.

    Google Scholar 

  20. Scanlon, T.M. 1998. What we owe to each other. Cambridge, MA: Harvard University Press.

    Google Scholar 

  21. Scanlon, T.M. 2013. Reply to Zofia Stemplowska. Journal of Moral Philosophy 10: 508–514.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Jussi Suikkanen.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Suikkanen, J. Ex Ante and Ex Post Contractualism: A Synthesis. J Ethics 23, 77–98 (2019). https://doi.org/10.1007/s10892-019-09282-6

Download citation


  • Contractualism
  • Social risk imposition
  • Normative ethics
  • Aggregation
  • Ex ante
  • Ex post
  • T.M. Scanlon