Dismantling Liberal Ethics: Why the Harm Principle Fails


In debates over controversial social issues, liberals often invoke a version of what philosophers call the harm principle. It goes like this: Adults should be legally permitted to do whatever they desire provided that they do not harm themselves or any other parties without the consent of all parties involved. The harm principle justified practices such as euthanasia, pornography, prostitution, drug use, and the like, because these involve only adults and do not seem to cause any non-consensual harms. And while it is a relatively novel concept, having only originated in the mid-nineteenth century, the harm principle is accepted today by most people—including many professed conservatives—as self-evident. But these people are wrong to accept it, because far from being self-evident, the harm principle, which undergirds most contemporary liberal ethics, proves quite implausible under close examination.

We see its weakness most clearly when we consider cases featuring practices that involve only adults and which do not cause non-consensual harms, but which every rational person should recognize as impermissible. Take, for instance, the following three:

(1) A film director is working on a gruesome scene that depicts the torture and subsequent execution of a man. In order to make the scene as realistic as possible, he finds a suicidal and masochistic adult who is willing to actually undergo both torture and execution. Both the film director and the adult consent to the plan of the former, and the adult is tortured and killed on the set.

(2) A man carrying a syringe and several vials of heroin and other similar substances approaches a friend and asks to be injected with the liquid in the vials. The man explains to his friend that, for reasons of his own, he wishes to become addicted to these substances for life but is too nervous to perform the injections himself. He further requests that his friend continue to inject him with the substances throughout his life so that he may never be productive to society again. The friend agrees and consensually executes the man’s plan.

(3) A cannibal places an advertisement on the Internet requesting an adult who is willing to be slaughtered and then eaten. An adult contacts the cannibal and consents to the request. The cannibal and the adult meet in person, and the adult is killed and consumed.

By all appearances, these cases involve only adults and do not involve non-consensual harms to anyone involved. Yet, I suspect that no rational person would agree that these may be ethically permitted. This effectively falsifies the harm principle. At this point, the liberal could make four objections to my reasoning, all of which merit careful attention.

Objection 1: The practices in these three cases are hypothetical, bear little resemblance to the ways in which people actually behave, and, therefore, are irrelevant to discussions about the harm principle.

The problem with this objection, however, is that it evinces a failure to understand why the cases were invoked in the first place. If we recall the original statement of the harm principle, we should notice that it makes a universal claim about the actions of consenting adults. Specifically, it says that adults should be legally permitted to do whatever they desire so long as they do not cause any non-consensual harms—period. The principle gives no qualifications that restrict the principle’s application to only “realistic” cases. So, as long as the practices in the three cases are humanly possible in principle (which they are), it is fair game, so to speak, to invoke them in criticizing the harm principle. Of course, the liberal making this objection is free to weaken the harm principle—perhaps by making it a non-universal claim that nevertheless is usually true—but once he does so, it is no longer clear that it will get him the conclusions about social issues that he wants. After all, if the liberal starts allowing for exceptions, then social conservatives are free to argue that practices like prostitution, which would have been legally permitted under the original harm principle, are also exceptions to the weakened harm principle. The liberal would then have to give some principled, non-arbitrary reason for thinking that the weakened harm principle still applies to the practices he wants it to, an endeavor whose results do not look promising. Moreover, an honest look at human beings’ great capacities for evil and depravity reveals that cases such as these are not even that unrealistic. In fact, the cannibalism case actually happened in Rotenburg, Germany, in 2001. So, even if we grant the liberal’s objection, it still fails to deal with at least one of the cases.

Objection 2: The victims involved in the cases were—despite appearances to the contrary—insane, and since sanity is necessary for genuine consent, the three cases therefore do not involve consenting adults.

This objection fails for similar reasons as Objection 1. While genuine consent does require sanity, there is no reason for thinking that the victims in the cases were insane. If the liberal peddling this reply says that “it’s just obvious” that the victims were insane, then why is the conservative prohibited from saying the same about those who consent to, say, euthanasia or drug use? Why can he not say that adults engaging in prostitution and drug use are—despite appearances to the contrary—insane? Again, the liberal objecting in this manner owes us some independent account that explains why the victims in the three cases were insane while prostitutes and drug users are not. Unless and until such an account is given, we have no reason for taking this objection seriously.

Objection 3: These cases involve practices that do cause non-consensual harms.

According to this objection, if we understand the concept of harm in a very broad sense so as to include psychological and potential harms, then we can say that uninvolved third parties who come to know about these practices are harmed in both ways. They are harmed psychologically by learning the disturbing details of these cases and are harmed potentially since these practices might motivate others to engage in violent behavior. Since the third parties clearly have not consented to these harms, the cases do not count as legitimate counterexamples to the harm principle. The obvious problem with this objection is that it proves too broad even for the liberal. If the concept of harm is going to be construed so sweepingly as to include psychological and potential harms, then it cannot be denied that practices supported by the liberal cause those harms as well. But, if that is true, then the harm principle cannot be employed for supporting their legalization.

Objection 4: The practices featured in the three cases should be legally permitted after all.

This is better described as a concession rather than an objection, as it simply accepts that the principles in these cases are ethically permissible. Shocking as it may be to some, this strategy appears to gained recent popularity among liberals. About a century ago, most people would have deemed these cases as decisive refutations of the harm principle, but modern liberal ethicists have proved less squeamish. As a result, a dialectical stalemate between the conservative and the liberal seems inevitable if the liberal makes this response.

Or perhaps not. While the concessionary liberal might be unpersuaded by the three cases, there might be other cases involving practices that even he would not deem legally permissible given his other ideological commitments. Let him grapple with the following cases:

(4) A white explorer journeys to central Africa and non-violently persuades a local tribe of Africans that his race is superior to theirs. From this, he argues that the Africans should become his slaves for life, raise their children to serve as slaves also, and return with him to his homeland to serve him there. For whatever reason, the members of the tribe agree with him and follow his recommendation.

(5) An adult who identifies as a homosexual wishes to undergo “treatment” for his condition. His treatment is both physically and psychologically painful, but he consents to both of these harms and begins a treatment program.
The practices described in these cases involve only consenting adults and do not involve non-consensual harms, but I suspect most liberals would rightly say that they should not be legally permitted. If my suspicion is correct, then there might be a way to break the deadlock after all.

There is one objection that I have not addressed, and that is the objection that none of the victims in any of my cases gave informed consent. It is informed consent, the liberal might claim, that is assumed in the harm principle, not just any old consent. Of course, in order for this objection to be substantially different from Objection 3, the liberal must be requiring more than sanity to be necessary for informed consent. But what might that be? I presume that the liberal thinks that informed consent requires acquaintance with the right facts, including normative facts about what ought to be done. If this is what is meant, though, then the question is no longer about consent, but rather about what the right facts are. And, if answering the question of what the right facts are calls for answering the question of what normative facts are, then social conservatives have their own intellectual resources upon which to draw from, especially the natural law tradition in ethics. So that is where the debate must take place. And, in my estimation, once arguments from natural law are allowed into fray, the liberal will be hard-pressed to adequately refute them.

Pranav Bethala is a sophomore from Gulfport, MS, majoring in the Philosophy Department. He can be reached at pbethala@princeton.edu.

Print Friendly, PDF & Email



Leave a Reply