The greatest ethical puzzle in philosophy?
[The Non-Identity Problem]
The interpretation of “harm” is well-grounded in ethics and philosophy, but it raises a problem coined in Derek Parfit’s 1984 book, Reasons and Persons, as the “non-identity problem”—a problem that goes unanswered by our accepted moral frameworks. One of the main points of contingency as to whether or not an action is “harmful” is the existence of an entity that is harmed—simply put, harm cannot exist in a vacuum; it necessitates that there must be something to apply the harm to, something that would be better off if the harmful action were not to be carried out.
Consider a moral conundrum involving two mothers, both with the intent to harm their children, and the means by which they harm their children involves taking a pill that guarantees that any child conceived/being conceived within two months of taking the pill will suffer from deafness for the rest of their lives.
The first mother is already eight months into pregnancy, bearing a well-developed fetus; so she takes it and ensures that the future child is deaf for the rest of his/her life. In this case, there is an individual who is made worse off by the mother’s actions—the baby could’ve been born as a perfectly healthy individual, but the actions of the mother 1 month before birth made it so that his/her quality of life has been degraded, or harmed. The future child still has a life worth living, but a life worse off than if his mother hadn’t taken the pill.
The second mother hasn’t even conceived of a baby yet, she hasn’t become pregnant yet, but also wants to harm her child, so she takes the pill. If she conceived two months after taking the pill, the effects would wear off, and any child born would be born perfectly as a perfectly healthy individual. Instead, the second mother conceives when the pill’s effects are still potent, making sure that the newborn will be deaf for the rest of his life—still a life worth living, but is there even a person being made worse off by this? If the child were to be born two months later, he would be a different person; the first child wouldn’t have existed in the first place. If the second mother gave the first individual a life worth living, albeit with an auditory impairment, who is even harmed?
It would be evidently stupid to say that the potential later child birthed after the pill’s effects wore off is harmed, because you can’t physically harm something that doesn’t physically exist at all—even then, you can’t “harm” potential children by not actualizing their potential to become a fetus, which has the potential to become an individual with a life worth living; that would mean any viable adult would be harming thousands of “individuals” every day by not having babies.
Can we do no wrong?
That hypothetical forms the basis of Parfit’s “non-identity problem”, which questions how actions that affect the identity of certain individuals or collectives can cause harm to them, therefore being morally wrong, when the individual or collective wouldn’t have existed otherwise. Not only does accepting the implications of this problem excuse actions that, traditionally, are unanimously agreed upon to be “harmful”; therefore, being morally wrong, but it also has moral implications for societies, for environmental policy, and for humanity as a whole.
As long as you give a certain population a “life worth living”, there can’t really be any objection raised to this; you would be charitable by bringing into existence individuals who can look forward to lives worth living. Consider another societal-scale hypothetical. Taking two mutually exclusive courses of action into consideration, we can either: a) conserve resources for future generations to maintain the stability of our planet, or b) use as many resources now, and make the next generation suffer, forcing them to sacrifice net prosperity so that humanity survives as a species.
It is rational to assume that with those drastic lifestyle changes, relocations to areas not subject to warming-induced sea level rise, and conflicts being fought over resources, the identities of all individuals within the next several generations will constitute different people. They will live lives far worse off than that of our generation, but nonetheless, they will have lives worth living. Due to harm needing an existent concept or individual to predicate the harm onto, there can now be no apparent justification for acting altruistically to preserve the well-being of the human species as a whole. It almost seems as if we can’t do any harm when making decisions that affect the individuality of something.
Logical forensics
Parfit’s non-identity problem relies on two premises to be true.
First, what is bad must be bad for someone. To add more nuance to this claim, “a choice makes a future morally worse only if at least one person is worse off because of that choice”.
Second, the existence of an identity is precarious—if you were conceived a second later, a different sperm cell would have fused to form a different zygote that would’ve evolved to form a different “you”, ergo any future-shaping decision cannot impact a person; any future-shaping decision impacts which person(s) become existent.
With those two simple premises, the paradox is completely assembled.
Potential solutions?
The most obvious potential solution to this problem lies in using a utilitarian framework that rejects the non-identity premise. When considering societal decisions, you compare the aggregate quality of life achieved by taking option a) to the aggregate quality of life achieved by taking option b), regardless of the fact that the samples contain different “identities” or individuals.
Utilitarian impersonalism resolves the main issue brought up by the non-identity problem; however, it fails to provide a holistic moral framework—you would have to accept the “repugnant conclusion” under this type of utilitarianism. Refer back to our hypothetical involving two different societies stemming from two different outcomes—if society a) gave 100 million people a life barely worth living, and society b) gave 10 million people a first-rate quality of life; assuming that the threshold of a life worth living is greater than or equal to 50% of a first-rate quality of life, society a) would still be five times “more happy” than society b), implying that it is ethically right to choose the former over the latter.
That doesn’t seem to make sense at all—having 100 million people living lives barely worth living would be reminiscent to some of the most corrupt, violent, and impoverished regions of the world, while having 10 million people living at an optimal standard of life would be comparable to a country like Sweden, ranked almost universally as on the podium for standard of living. That can’t be the answer—and even if it were plausible, we can definitely do better than that!
Or you could accept the conclusion. The actions are not wrong after all, and the non-identity problem simply brings up an uncomfortable truth about normative ethics. Actions like purposefully giving birth to a deaf child are definitely disturbing, but not really wrong. This is probably the most unintuitive, and by far the most gratuitous, imprudent solution—so most philosophers reject it.
Both of these solutions seem to logically “soft-lock” you, but don’t fret—plenty of other moral interpretations provide plausible solutions to this problem.
You can wrong someone without making them worse off! There are many intrinsic moral values and rights that can be acquired ontologically, and violating any of these intrinsic values is harm done to someone. People have the right to be free moral actors; they have the right to not ontological harm, along with a multitude of other prerogatives that are contingent on being human. With consideration to Kantian ethics, not respecting a person’s morality is also morally wrong; moral harm is not one-dimensional.
Comparing descriptions and relations is better than treating existence as precarious. Non-identity is not a valid theory—if you were comparing a child that was conceived by the same couple, you would refer to it as “the couple’s child”, regardless of which specific sperm and egg cell fused to create it, how it developed during pregnancy over 9 months, or how the child is raised. A quite intuitive rationalization of this would lie within the fluidity of our own identities. You also shape your own identity with every single choice you make; if you went to school a) instead of school b) or ate egg tart a) instead of egg tart b), your life would be entirely different in 30 years, meaning if you accepted the paradox, you wouldn’t be able to apply any moral principles as identity is always in a state of flux.
Use probability theory! The paradox implies that if your parents had acted even a bit differently in the process of your conception, you wouldn’t exist. You can cross-apply that same logic to the concept of non-identity.
Choice e) creates a person x) who has a difficult but worthwhile life.
Choice f) is better, but produces a different person, y).
The paradox assumes that person x) is much more likely than person y) under the harmful choice. The existence of any specific person is already astronomically minuscule; there is no dichotomy between choice e), person x) and choice f), person y). Even if choice e) was chosen during the conception process, thousands of sperm would still be milliseconds away from actualizing one of their thousands of possible unique identities—person x) is not guaranteed at all.
Wrongdoing lies within immoral intent. A mother who wants to give their child an impairment is morally wrong because she is bringing someone into life with foreseeable, intentionally added suffering.
Despite the proposed solutions, Parfit’s non-identity problem still stands as one of the largest puzzles within the field of moral philosophy. It makes us rethink even the most foundational assumptions of harm-based morality; Parfit himself stated that a holistic fix to this problem necessitates “Theory X”, an entirely new moral framework.
I hope you enjoyed reading this article as much as I enjoyed writing it—before you go, which of the 4 alternative solutions sounds the most plausible?
This really made me think. From an Indigenous perspective, we’re taught to consider the impact of our actions on the next seven generations. It shifts the question away from “who is harmed?” to “what kind of world are we choosing to create for those who come after us?” Even if identity changes, the responsibility doesn’t.
Haven’t read something mind wringing in quite some time. Pretty sure i’ll have more to say or absorb on the matter on a second read some time in future 👏