Skip to main content
Log in

Asymmetric population axiology: deliberative neutrality delivered

  • Published:
Philosophical Studies Aims and scope Submit manuscript


Two related asymmetries have been discussed in relation to the ethics of creating new lives: First, we seem to have strong moral reason to avoid creating lives that are not worth living, but no moral reason to create lives that are worth living. Second, we seem to have strong moral reason to improve the wellbeing of existing lives, but, again, no moral reason to create lives that are worth living. Both asymmetries have proven very difficult to account for in any coherent moral framework. I propose an impersonal population axiology to underpin the asymmetries, which sidesteps the problematic issue of whether or not people can be harmed or benefited by creation or non-creation. This axiology yields perfect asymmetry from a deliberative perspective, in terms of expected value. The axiology also yields substantial asymmetry for large and realistic populations in terms of their actual value, beyond deliberative relevance.

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

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others


  1. McMahan does not endorse these asymmetries. His treatment is more exploratory than argumentative. He does tend to group lives that will exist in the future with presently existing lives, drawing a line between these two classes on the one hand and on the other lives that may or may not exist in the future (e.g. 2013, 11–12). I will not distinguish between future lives that will and future lives that may or may not exist. One reason not to is that from a collective, deliberative perspective, there are no future lives that will exist regardless of what we do—any future life can be prevented and is in that sense contingent.

  2. Lifetime wellbeing is the only sort of wellbeing I will consider throughout and so I will drop the "lifetime" descriptor in the following. I will not discuss the nature of wellbeing, but I will assume that it is interpersonally comparable.

  3. My discussion is not necessarily restricted to people, or to human lives, though most of the previous debate on these matters has been framed in terms of human wellbeing and human lives.

  4. For an recent and unusually comprehensive defense of nativism, se Ord (2014).

  5. Two significant recent normative defenses of the first asymmetry, which also contain ample references to earlier attempts, are Roberts (2011) and Algander (2013).

  6. Recent significant examples include Hare (2007) and Arrhenius (2009) (the latter is more an investigation than an outright defense).

  7. In a recent book, Broome even identifies neutrality with a version of the person-affecting restriction: "The [neutrality] intuition is that, when something changes in the world, we can evaluate the change on the basis of how good it is for the people who exist" (2012, 171).

  8. Broome recognizes that these are possible goods early in Weighing Lives (2004, 43–45).

  9. Narveson appears not to have realized how problematic it is to hold both that positive additions are neutral and, as he did, that a more positive addition is better than a less positive one. He claims that "if you are going to produce people, then you should produce the happiest ones consistently with the equal consideration of all others concerned", but fails to notice that this claim is in strong tension with his other views (1978, 53). As he goes on to say, "the utility of those to be produced need to be consulted only if they actually are produced" (ibid.). But then you cannot, as instructed in the first qoute, consult the relative happiness of various groups of potential people in order to determine who are the happiest.

    In his most recent refutation of axiological neutrality, Broome has modified his argument, perhaps in order to account for the possible value of equal distribution of wellbeing (2012, 176–177). Since I am not concerned with such equality, and since I find the modified argument less clear and less convincing, I stay with the earlier version.

  10. Principles of this sort invoked in population axiology include the strong Pareto axiom of Blackorby et al. (2005, 69) and Gustaf Arrhenius’ Pareto Dominance (2013, 213).

  11. As noted by Persson (1997), Feldman's presentation oscillates between different ideas on how to factor in desert. However, Arrhenius (2013, chapter 8) has reconstructed Feldman's view such that it is unambiguous and consistent.

  12. It will be imperative to understanding my view to remember the distinction between value and wellbeing, as well as the distinction between either of these being either expected or actual.

  13. I take no stand on the controversial issue of when exactly a life starts. In other words, I follow Narveson in "neglecting the question about the point at which a person comes into existence" (1967, 63).

    One could introduce an upper limit for critical levels in addition to the lower limit. However, I find this restriction less intuitive than the lower limit. Also, there is no obvious level of wellbeing at which to fix it. Therefore, and to simplify my presentation, I will ignore this possibility in the following. It should be kept in mind, however, especially by those who are somewhat inclined towards the total view, since expectism with an upper limit on critical levels approaches the total view as this upper limit approaches zero wellbeing. In fact, expectism with a relatively low upper limit behaves much like a standard critical level view, while it avoids, from the deliberative perspective, the counter-intuitive implication that lives with positive wellbeing can have negative value.

  14. In other words, expectism is consistent with what Broome calls separability of lives. Broome derives separability of lives from the principle of personal good, but it has independent appeal. Notably, the average view is inconsistent with separability of lives, since it implies that the value of any change in the wellbeing of an existing life is relative to the size of the population (individual changes are more important the smaller the population). It seems to me that the intuition that the value of a life should not depend on population size is a close cousin to neutrality: If a population is not improved by positive additions, then neither should these additions affect the value of changes in the wellbeing of existing people.

  15. If one adopts, as does Broome, an expected value view of the normative import of consequences (whether or not one is a consequentialist or only considers consequences one morally relevant factor), then this deliberative perspective is also the only normatively relevant perspective.

  16. Note that I am here referring to objective probability, while I previously referred to an objective perspective in the sense of all-knowing, retrospective or actualist. The objective probability of throwing a 6 on a standard dice may be 1/6, but the actual outcome either is 6 or is not 6.

  17. Nor of course could non-human beings, unless perhaps some advanced alien or technological intelligences that come into being highly developed.

  18. Idealized subjective probabilities could for example be such as would be attributed by a fully rational team of diverse scientists with unlimited time and access to all information collectible with the best current (or future) scientific methods as well as the best methods for analyzing that information.

    It is no threat to expectism that subjective probabilities are idealized to the point that they are perfect predictions, such as could perhaps be made by an omniscient being. In fact, this would just help underpin neutrality. However, critical levels must not be defined directly as actual outcomes, since this would imply that increased wellbeing has no value.

    Note that with objective probabilities, the deliberative perspective is not the subjective perspective. A person that has not accepted expectism may reasonably but falsely believe that the expected value of a life she creates is positive and so be subjectively justified in creating it. Any person that accepts and understands expectism, on the other hand, will know that the expected value of any future life is zero or negative, purely on theoretical grounds.

  19. Prioritarianism cannot be incorporated into expectism as straightforwardly as for totalism or averagism, but it can be added as a separate element, such that the value of a life is the divergence of its actual wellbeing from its critical level, plus its priority value. To maintain neutrality, the priority value must be based only on negative and not on positive wellbeing (the simplest version would be to let it be identical to negative wellbeing).

  20. Compare Arrhenius argument that the distinction between lives which are good to live and lives which are not is substantial and “has nothing to do with numbers”. We can represent the distinction between these two types of lives by the value zero on a scale, but this does not mean that the distinction is merely an arbitrary number (Arrhenius 2013, 38–39).

  21. These views are, like the average view and unlike expectism, inconsistent with separability of lives. Cf. footnote 14.

  22. Arrhenius (2013, 278) calls this view “asymmetric presentism”.

  23. The negative repugnant conclusion closely resembles what Arrhenius (2000) has called “the sadistic conclusion”—that adding lives with negative wellbeing to a population can be better than adding lives with positive wellbeing.


  • Algander, P. (2013). Harm, benefit, and non-identity. Dissertation, Uppsala University.

  • Arrhenius, G. (2000). An impossibility theorem for welfarist axiologies. Economics and Philosophy, 16, 247–266.

    Article  Google Scholar 

  • Arrhenius, G. (2009). Can the person affecting restriction solve the problems in population ethics? In M. A. Roberts & D. T. Wasserman (Eds.) Harming future persons (pp. 289–314). International Library of Ethics, Law, and the New Medicine 35. Springer Netherlands.

  • Arrhenius, G. (2013). Population ethics: The challenge of future generations. Manuscript.

  • Blackorby, C., Bossert, W., & Donaldson, D. (2005). Population issues in social choice theory, welfare economics, and ethics. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Bradley, B. (2013). Asymmetries in benefiting, harming and creating. The Journal of Ethics, 17, 37–49.

    Article  Google Scholar 

  • Broome, J. (1999). Ethics out of economics. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  • Broome, J. (2004). Weighing lives. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Broome, J. (2005). Should we value population? Journal of Political Philosophy, 13, 399–413.

    Article  Google Scholar 

  • Broome, J. (2012). Climate matters: Ethics in a warming world. New York: W. W. Norton & Company.

    Google Scholar 

  • Feldman, F. (1995). Justice, desert, and the repugnant conclusion. Utilitas, 7(2), 189–206.

    Article  Google Scholar 

  • Hare, C. (2007). Voices from another world: Must we respect the interests of people who do not, and will never, exist? Ethics, 117, 498–523.

    Article  Google Scholar 

  • Holtug, N. (2010). Persons, interests, and justice. Oxford: Oxford University Press.

    Book  Google Scholar 

  • McMahan, J. (1981). Problems of population theory. Ethics, 92, 96–127.

    Article  Google Scholar 

  • McMahan, J. (2009). Asymmetries in the morality of causing people to exist. In M. A. Roberts & D. T. Wasserman (Eds.) Harming future persons (pp. 49–68). International Library of Ethics, Law, and the New Medicine 35. Springer Netherlands.

  • McMahan, J. (2013). Causing people to exist and saving people’s lives. The Journal of Ethics, 17, 5–35.

    Article  Google Scholar 

  • Narveson, J. (1967). Utilitarianism and new generations. Mind, New Series, 76(301), 62–72.

    Article  Google Scholar 

  • Narveson, J. (1973). Moral problems of population. The Monist, 57, 62–66.

    Article  Google Scholar 

  • Narveson, J. (1978). Future people and us. In R. I. Sikora & B. Barry (Eds.), Obligations to future generations. Philadelphia: Temple University Press.

    Google Scholar 

  • Ord, T. (2014). Overpopulation or underpopulation? In I. Goldin (Ed.), Is the planet full?. Oxford: Oxford University Press.

    Google Scholar 

  • Parfit, D. (1984). Reasons and persons. Oxford: Oxford University Press.

    Google Scholar 

  • Persson, I. (1997). Ambiguities in Feldman’s Desert-adjusted values. Utilitas, 9(3), 319–328.

    Article  Google Scholar 

  • Roberts, M. A. (2011). The asymmetry: A solution. Theoria, 77, 333–367.

    Article  Google Scholar 

  • Temkin, L. S. (2003). Equality, priority or what? Economics and Philosophy, 19(1), 61–87.

    Article  Google Scholar 

Download references


The core aspects of the view developed in this paper were first presented at a workshop on Climate change policy after Copenhagen, in Uppsala in February, 2010. Versions have since been presented at the philosophy departments in Uppsala and Umeå, as well as at the ISUS conference in Yokohama 2014. I am grateful for the feedback received. Lars Lindblom and Niklas Möller provided helpful criticism of an early draft. Per Algander, Erik Carlson and Stephen Wilkinson each did the same for two separate drafts. My wife Camilla Grill asked helpful, probing questions based on my many animated explanations. Last but certainly not least, a reviewer for this journal provided very helpful comments.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Kalle Grill.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Grill, K. Asymmetric population axiology: deliberative neutrality delivered. Philos Stud 174, 219–236 (2017).

Download citation

  • Published:

  • Issue Date:

  • DOI: