The Non-Identity Problem
If a policy choice is necessary for a person's existence, how can that person have been harmed by it, even if the policy left them worse off?
Derek Parfit identified this problem in 1984. It threatens to undermine our ability to hold governments morally responsible for policies with long-term harmful consequences, because the people who will bear those consequences owe their very existence to the policies that caused them.
Parfit, D. (1984). Reasons and Persons. Oxford University Press.
The problem
A government chooses a policy of rapid resource extraction. It knows the policy will deplete resources significantly, leaving people born 200 years from now with a lower quality of life than they would have had under a more conservative policy.
But here is the catch: if the government had chosen the alternative policy, different people would have been born. The choices made over those 200 years, who worked where, who met whom, who had children and when, would all have been different. The people who will actually exist under the depleting policy owe their existence to that policy.
Standard accounts of harm require that someone be made worse off relative to some baseline. But the future people cannot compare their lives to the alternative. In the alternative, they would not have existed. Their lives, even if difficult, may well be worth living. They have not been harmed in the standard sense.
Why this seems to let governments off the hook
The non-identity problem appears to grant a perverse kind of immunity. Any policy that affects who gets born, even a reckless one, cannot be said to harm the people who result from it, as long as their lives are worth living.
This seems plainly wrong. A government that knowingly impoverishes future generations should be morally accountable. But the standard person-affecting framework, which asks "who is worse off than they would otherwise have been?", gives no satisfying answer when the affected people would not otherwise have existed at all.
Parfit's response
Parfit argued that the non-identity problem exposes a deep flaw in person-affecting principles, the view that outcomes only matter insofar as they make particular people better or worse off.
His proposed alternative: impersonal principles that evaluate outcomes based on how much well-being exists, without anchoring the evaluation to the interests of specific individuals. On this view, it matters that a policy produces a worse world, even if no particular person can claim to have been wronged.
This move has its own costs. Impersonal principles tend to lead toward the Repugnant Conclusion. The non-identity problem does not have a tidy solution. It is evidence that our moral concepts were not built to handle the full range of questions we now need them to answer.
Discussion questions
- If you decided to have children knowing they would face serious hardship, did you harm them?
- Can a policy wrong future people if those exact people would not have existed under the better policy?
- How should governments make decisions that affect people who do not yet exist?
Take it to the dinner table.
Get 3 thought experiments for memorable conversations, designed for dinner, with friends, at events, or anywhere small talk has gone on too long.